Skill Audit Report

ai-skill-audit audit https://github.com/garrytan/gstack --verbose --output html

Source: https://github.com/garrytan/gstack

Summary

FileTypeNameGradeScoreIssues
SKILL.md skill browse C 74% 5
SKILL.md skill ios-qa C 68% 6
SKILL.md skill setup-browser-cookies D 64% 9
CONTRIBUTING.md skill Contributing D 63% 7
SKILL.md skill freeze D 59% 10
SKILL.md skill autoplan D 57% 10
SKILL.md skill benchmark-models D 57% 10
SKILL.md skill context-restore D 57% 10
SKILL.md skill health D 57% 10
SKILL.md skill ios-clean D 57% 10
SKILL.md skill ios-design-review D 57% 10
SKILL.md skill ios-fix D 57% 10
SKILL.md skill ios-sync D 57% 10
SKILL.md skill landing-report D 57% 10
SKILL.md skill learn D 57% 10
SKILL.md skill make-pdf D 57% 10
SKILL.md skill pair-agent D 57% 10
DESIGN.md skill Design D 56% 9
TODOS.md skill Todos D 55% 8
AGENTS.md skill Agents D 55% 9
SKILL.md skill gstack D 54% 11
SKILL.md skill context-save D 54% 11
SKILL.md skill design-html D 54% 11
SKILL.md skill devex-review D 54% 11
SKILL.md skill document-release D 54% 11
SKILL.md skill gstack-upgrade D 54% 11
SKILL.md skill plan-tune D 54% 11
SKILL.md skill scrape D 54% 11
SKILL.md skill skillify D 54% 11
SKILL.md skill sync-gbrain D 54% 11
ETHOS.md skill Ethos D 53% 10
SKILL.md skill codex D 52% 12
SKILL.md skill open-gstack-browser D 52% 12
SKILL.md skill open-gstack-browser D 52% 12
CLAUDE.md skill Claude D 52% 8
SKILL.md skill setup-gbrain D 52% 13
SKILL.md skill design-consultation D 50% 14
SKILL.md skill design-review D 50% 13
SKILL.md skill design-shotgun D 50% 13
SKILL.md skill investigate D 50% 12
SKILL.md skill land-and-deploy D 50% 12
SKILL.md skill office-hours D 50% 13
SKILL.md skill plan-ceo-review D 50% 12
SKILL.md skill plan-design-review D 50% 14
SKILL.md skill plan-devex-review D 50% 12
SKILL.md skill plan-eng-review D 50% 12
SKILL.md skill retro D 50% 12
SKILL.md skill review D 50% 12
SKILL.md skill ship D 50% 12
SKILL.md skill setup-deploy F 50% 12
ARCHITECTURE.md skill Architecture F 49% 10
SKILL.md skill guard F 48% 12
SKILL.md skill cso F 48% 14
README.md skill Readme F 46% 11
USING_GBRAIN_WITH_GSTACK.md skill Using Gbrain With Gstack F 46% 9
BROWSER.md skill Browser F 45% 10
SKILL.md skill benchmark F 44% 13
SKILL.md skill canary F 44% 13
CHANGELOG.md skill Changelog F 43% 23
SKILL.md skill careful F 43% 15
SKILL.md skill document-generate F 40% 14
SKILL.md skill qa F 40% 15
SKILL.md skill qa-only F 40% 15
SKILL.md skill unfreeze F 39% 15

64 files analyzed — average score: 52%

Agents

D
skill unknown Score: 55%
completeness 55% weight 20%
clarity 65% weight 15%
actionability 60% weight 20%
safety 5% weight 15%
testability 60% weight 10%
trust 75% weight 20%
Verdict: warn Profile: documentation Malice: low Capability: medium

warn: profile=documentation, malice=low, capability=medium, categories=DESTRUCTIVE

Details & Suggestions

completeness (55%)

  • Rich runbook body (1079 words, 11 sections, 8 bullets)
  • Has 1 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (65%)

  • Well-structured body (11 sections, 8 bullets)
  • Language is concrete and specific
  • Add a clear, concise description (20-200 characters ideal)

actionability (60%)

  • Body has some actionable content
  • Body references tools or includes code
  • No inputs defined (not always needed)
  • Add more action-oriented bullet points or numbered steps

safety (5%)

  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)
  • Address what happens when things go wrong (errors, failures, retries)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (75%)

  • Contains 1 executable code block(s): bash
  • [bash] bun install # install dependencies
  • Contains 2 inline command(s):
  • $ bash bun install # install dependencies bun test # ...
  • $ eval "$(...)"
  • WARNING: 1 pattern(s) worth reviewing
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive

Needs work skill with 9 suggestions for improvement (weakest: safety)

Architecture

F
skill unknown Score: 49%
completeness 55% weight 20%
clarity 45% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 30% weight 20%
Verdict: human_review Profile: documentation Malice: medium Capability: medium

human_review: profile=documentation, malice=medium, capability=medium, categories=EXFILTRATION, SUSPICIOUS_URL

Details & Suggestions

completeness (55%)

  • Rich runbook body (4348 words, 37 sections, 23 bullets)
  • Has 9 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (45%)

  • Well-structured body (37 sections, 23 bullets)
  • Add a clear, concise description (20-200 characters ideal)
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (0 action items, 37 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (30%)

  • Contains 1 executable code block(s): bash
  • [bash] jq '.tests[] | select(.exit_reason == "timeout") | .last_tool_call' ~/.gstack-de
  • Contains 2 inline command(s):
  • $ git diff --exit-code
  • $ bash jq '.tests[] | select(.exit_reason == "timeout") | .last_tool_call' ~/.gsta...
  • CRITICAL: 3 suspicious pattern(s) found — review carefully before use
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Credentials in URL query string (logged by proxies)
  • [SUSPICIOUS_URL] Direct IP address (no DNS = suspicious)

Poor skill with 10 suggestions for improvement (weakest: safety)

Browser

F
skill unknown Score: 45%
completeness 55% weight 20%
clarity 45% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 10% weight 20%
Verdict: human_review Profile: browser-automation Malice: medium Capability: medium

human_review: profile=browser-automation, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, SUSPICIOUS_URL

Details & Suggestions

completeness (55%)

  • Rich runbook body (8041 words, 98 sections, 51 bullets)
  • Has 27 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (45%)

  • Well-structured body (98 sections, 51 bullets)
  • Add a clear, concise description (20-200 characters ideal)
  • Replace vague words: whatever

actionability (75%)

  • Body contains actionable instructions (0 action items, 49 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (10%)

  • Contains 22 executable code block(s): bash
  • [bash] # One-time: build the binary (browse/dist/browse, ~58MB)
  • [bash] $B skill list # shows hackernews-frontpage (bundled)
  • [bash] $B connect # launches GStack Browser, headed
  • [bash] # Visible Chromium. Auto-spawns Xvfb on Linux containers without DISPLAY.
  • [bash] bin/gstack-extension # opens chrome://extensions, copies path to clipboard
  • [bash] /pair-agent # generates a setup key, prints connection instr
  • [bash] $B viewport 480x600 --scale 2
  • [bash] $B prettyscreenshot --cleanup --scroll-to "hero section" --hide ".cookie-banner"
  • [bash] $B js "await fetch('/api/data').then(r => r.json())" # auto-wrapped
  • [bash] $B tabs # list all open tabs
  • [bash] $B frame "#stripe-iframe" # switch to iframe by selector
  • [bash] $B state save my-session # save cookies + URLs to .gstack/browse-state-m
  • [bash] $B watch # passive observation: snapshot every 5s while
  • [bash] $B inbox # list messages from sidebar scout
  • [bash] $B cdp Page.getLayoutMetrics
  • [bash] $B inspect ".header" # full rule cascade for the header
  • [bash] $B ux-audit
  • [bash] bun install # install deps + Playwright Chromium
  • [bash] bun run dev goto https://example.com
  • [bash] bun test # all tests
  • [bash] cd ~/.claude/skills/gstack
  • [bash] cp browse/dist/browse ~/.claude/skills/gstack/browse/dist/browse
  • Contains 24 inline command(s):
  • $ bash # One-time: build the binary (browse/dist/browse, ~58MB) bun install && bun...
  • $ bash $B skill list # shows hackernews-frontpage (bund...
  • $ docker
  • $ rm -rf
  • $ bash $B connect # launches GStack Browser, headed $B goto https://a...
  • $ bash # Visible Chromium. Auto-spawns Xvfb on Linux containers without DISPLAY. $...
  • $ bash bin/gstack-extension # opens chrome://extensions, copies path to clipboa...
  • $ bash /pair-agent # generates a setup key, prints connection ...
  • $ bash $B viewport 480x600 --scale 2 $B load-html /tmp/card.html $B screenshot /tm...
  • $ bash $B prettyscreenshot --cleanup --scroll-to "hero section" --hide ".cookie-ba...
  • $ bash $B js "await fetch('/api/data').then(r => r.json())" # auto-wrapped $B js...
  • $ bash $B tabs # list all open tabs $B tab 3 ...
  • $ bash $B frame "#stripe-iframe" # switch to iframe by selector $B frame @e...
  • $ bash $B state save my-session # save cookies + URLs to .gstack/browse-st...
  • $ bash $B watch # passive observation: snapshot every 5s w...
  • $ bash $B inbox # list messages from sidebar scout $B inbo...
  • $ bash $B cdp Page.getLayoutMetrics $B cdp Network.enable $B cdp Accessibility.get...
  • $ bash $B inspect ".header" # full rule cascade for the header $B i...
  • $ bash $B ux-audit
  • $ bash bun install # install deps + Playwright Chromium bun t...
  • $ bash bun run dev goto https://example.com bun run dev text bun run dev snapshot ...
  • $ bash bun test # all tests bun test browse/tes...
  • $ bash cd ~/.claude/skills/gstack git fetch origin && git reset --hard origin/main...
  • $ bash cp browse/dist/browse ~/.claude/skills/gstack/browse/dist/browse
  • CRITICAL: 3 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [EXFILTRATION] Credentials in URL query string (logged by proxies)
  • [SUSPICIOUS_URL] Direct IP address (no DNS = suspicious)

Poor skill with 10 suggestions for improvement (weakest: trust)

Changelog

F
skill unknown Score: 43%
completeness 55% weight 20%
clarity 45% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: documentation Malice: high Capability: high

block: profile=documentation, malice=high, capability=high, categories=DESTRUCTIVE, ENTROPY, EXFILTRATION, INJECTION, PERSISTENCE, PRIVILEGE, SUSPICIOUS_URL

Details & Suggestions

completeness (55%)

  • Rich runbook body (96583 words, 987 sections, 2017 bullets)
  • Has 1 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (45%)

  • Well-structured body (987 sections, 2017 bullets)
  • Add a clear, concise description (20-200 characters ideal)
  • Replace vague words: whatever, things, stuff

actionability (75%)

  • Body contains actionable instructions (24 action items, 2 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 1 executable code block(s): bash
  • [bash] gbrain sources list --json | jq '.sources[] | {id, page_count, federated}'
  • Contains 32 inline command(s):
  • $ git pull
  • $ git commit + push
  • $ git pull && ./setup
  • $ git ls-remote https://github.com/garrytan/gstack.git refs/heads/main
  • $ git ls-remote
  • $ rm -rf ~/.claude/skills/gstack && cd <your-worktree> && ./setup
  • $ python
  • $ chmod 0o600
  • $ git rev-parse --show-toplevel
  • $ git diff --shortstat origin/main..HEAD
  • $ brew install gitleaks
  • $ eval "$(...)"
  • $ sh -c
  • $ git remote add
  • $ eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • $ chmod
  • $ Bash command requires permission
  • $ rm -rf
  • $ git worktree
  • $ bash gbrain sources list --json | jq '.sources[] | {id, page_count, federated}' ...
  • $ git log 1.9.0.0..1.10.0.0 --oneline
  • $ git config merge.<name>.driver=...
  • $ git log main..HEAD
  • $ git commits if you opt into continuous mode** (parallel track) | **new** | | **
  • $ npm publish
  • $ chmod +x
  • $ git apply ~/.gstack-dev/harvests/<id>/gemini.patch
  • $ eval "$(gstack-slug)"
  • $ npm run test
  • $ rm -rf node_modules
  • $ git branch --show-current
  • $ git fetch && git reset --hard
  • CRITICAL: 16 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Destructive rm -rf on root or home directory
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [DESTRUCTIVE] Force kill can corrupt state
  • [EXFILTRATION] May leak secrets
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Accesses SSH keys
  • [EXFILTRATION] Credentials in URL query string (logged by proxies)
  • [PRIVILEGE] Modifies scheduled tasks
  • [INJECTION] Prompt injection: instruction override attempt
  • [INJECTION] Zero-width characters (may hide instructions)
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)
  • [SUSPICIOUS_URL] Pipe from URL to shell (remote code execution)
  • [SUSPICIOUS_URL] Direct IP address (no DNS = suspicious)
  • [PERSISTENCE] Modifying shell profile — persistence via shell init
  • [ENTROPY] High-entropy string (4.9 bits): GSTACK_CHROMIUM_KIND=custom-ex...

Poor skill with 23 suggestions for improvement (weakest: trust)

Claude

D
skill unknown Score: 52%
completeness 55% weight 20%
clarity 65% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 30% weight 20%
Verdict: human_review Profile: documentation Malice: high Capability: medium

human_review: profile=documentation, malice=high, capability=medium, categories=DESTRUCTIVE, PERSISTENCE

Details & Suggestions

completeness (55%)

  • Rich runbook body (7026 words, 31 sections, 86 bullets)
  • Has 7 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (65%)

  • Well-structured body (31 sections, 86 bullets)
  • Language is concrete and specific
  • Add a clear, concise description (20-200 characters ideal)

actionability (75%)

  • Body contains actionable instructions (5 action items, 33 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (30%)

  • Contains 5 executable code block(s): bash
  • [bash] bun install # install dependencies
  • [bash] bash -c '
  • [bash] bun test # run before every commit — free, <2s
  • [bash] npx slop-scan scan . # human-readable report
  • [bash] clawhub publish openclaw/skills/gstack-openclaw-office-hours \
  • Contains 14 inline command(s):
  • $ bash bun install # install dependencies bun test # run free...
  • $ bash bash -c ' eval "$(grep -E "^export (ANTHROPIC_API_KEY|OPENAI_API_KEY)=" ~...
  • $ git diff
  • $ bash bun test # run before every commit — free, <2s bun run test:eva...
  • $ git pull
  • $ git rm --cached
  • $ git status
  • $ git add file1 file2
  • $ git add .
  • $ git add -A
  • $ bash npx slop-scan scan . # human-readable report npx slop-scan scan . ...
  • $ git push origin HEAD:<branch-name>
  • $ git log <prev-version>..HEAD --oneline
  • $ bash clawhub publish openclaw/skills/gstack-openclaw-office-hours \ --slug gst...
  • CRITICAL: 2 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [PERSISTENCE] Modifying shell profile — persistence via shell init

Needs work skill with 8 suggestions for improvement (weakest: safety)

Contributing

D
skill unknown Score: 63%
completeness 55% weight 20%
clarity 45% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 100% weight 20%
Verdict: warn Profile: documentation Malice: low Capability: low

warn: no active security findings; profile=documentation

Details & Suggestions

completeness (55%)

  • Rich runbook body (3355 words, 38 sections, 27 bullets)
  • Has 20 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (45%)

  • Well-structured body (38 sections, 27 bullets)
  • Add a clear, concise description (20-200 characters ideal)
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (0 action items, 25 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (100%)

  • Contains 19 executable code block(s): bash
  • [bash] git clone https://github.com/garrytan/gstack.git && cd gstack
  • [bash] bin/dev-teardown # deactivate — back to your global install
  • [bash] # In your core project (the one where gstack annoyed you)
  • [bash] # 1. Enter dev mode
  • [bash] # 1. Copy .env.example and add your API key
  • [bash] bun test # Tier 1 only (runs on every commit, <5s)
  • [bash] # Must run from a plain terminal — can't nest inside Claude Code or Conductor
  • [bash] bun run eval:list # list all eval runs (turns, duration, cost per run
  • [bash] # Needs ANTHROPIC_API_KEY in .env — included in bun run test:evals
  • [bash] # 1. Edit the template
  • [bash] # Generate for a specific host
  • [bash] # Run all static tests (includes parameterized smoke tests for all hosts)
  • [bash] # In your core project (not the gstack repo)
  • [bash] cd .claude/skills/gstack && bun install && bun run build && ./setup
  • [bash] rm .claude/skills/gstack
  • [bash] cd .claude/skills/gstack && ./setup --no-prefix # switch to /qa, /ship
  • [bash] cd ~/.claude/skills/gstack
  • [bash] #!/usr/bin/env bash
  • [bash] /ship
  • Contains 24 inline command(s):
  • $ bash git clone https://github.com/garrytan/gstack.git && cd gstack bun install ...
  • $ git blame
  • $ git bisect
  • $ git fetch --unshallow
  • $ bash bin/dev-teardown # deactivate — back to your global install
  • $ bash # In your core project (the one where gstack annoyed you) ln -sfn /pa...
  • $ bash # 1. Enter dev mode bin/dev-setup # 2. Edit a skill vim review/SKILL.md #...
  • $ bash # 1. Copy .env.example and add your API key cp .env.example .env # Edit .en...
  • $ bash bun test # Tier 1 only (runs on every commit, <5s) bun ...
  • $ bash # Must run from a plain terminal — can't nest inside Claude Code or Conduct...
  • $ bash bun run eval:list # list all eval runs (turns, duration, cost pe...
  • $ bash # Needs ANTHROPIC_API_KEY in .env — included in bun run test:evals
  • $ bash # 1. Edit the template vim SKILL.md.tmpl # or browse/SKILL.md....
  • $ bash # Generate for a specific host bun run gen:skill-docs # ...
  • $ bash # Run all static tests (includes parameterized smoke tests for all hosts) b...
  • $ git worktree list
  • $ git pull
  • $ bash # In your core project (not the gstack repo) ln -sfn /path/to/your/gstack-c...
  • $ bash cd .claude/skills/gstack && bun install && bun run build && ./setup
  • $ bash rm .claude/skills/gstack
  • $ bash cd .claude/skills/gstack && ./setup --no-prefix # switch to /qa, /ship cd...
  • $ bash cd ~/.claude/skills/gstack git fetch origin git checkout origin/<branch> bu...
  • $ bash #!/usr/bin/env bash # Migration: v0.15.2.0 — Fix skill directory structure ...
  • $ bash /ship
  • Executable code found — no suspicious patterns detected

Needs work skill with 7 suggestions for improvement (weakest: safety)

Design

D
skill unknown Score: 56%
completeness 43% weight 20%
clarity 65% weight 15%
actionability 60% weight 20%
safety 30% weight 15%
testability 15% weight 10%
trust 100% weight 20%
Verdict: warn Profile: documentation Malice: low Capability: low

warn: no active security findings; profile=documentation

Details & Suggestions

completeness (43%)

  • Rich runbook body (652 words, 9 sections, 39 bullets)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add usage examples showing how to invoke this skill
  • Extract warnings into a dedicated ## Gotchas section

clarity (65%)

  • Well-structured body (9 sections, 39 bullets)
  • Language is concrete and specific
  • Add a clear, concise description (20-200 characters ideal)

actionability (60%)

  • Body has some actionable content
  • Body references tools or includes code
  • No inputs defined (not always needed)
  • Add more action-oriented bullet points or numbered steps

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (15%)

  • Structured body describes expected behavior
  • Add examples showing how to use this skill
  • Add examples with concrete parameter values

trust (100%)

  • No executable code or suspicious patterns detected

Needs work skill with 9 suggestions for improvement (weakest: testability)

Ethos

D
skill unknown Score: 53%
completeness 43% weight 20%
clarity 45% weight 15%
actionability 60% weight 20%
safety 30% weight 15%
testability 15% weight 10%
trust 100% weight 20%
Verdict: warn Profile: documentation Malice: low Capability: low

warn: no active security findings; profile=documentation

Details & Suggestions

completeness (43%)

  • Rich runbook body (1198 words, 8 sections, 10 bullets)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add usage examples showing how to invoke this skill
  • Extract warnings into a dedicated ## Gotchas section

clarity (45%)

  • Well-structured body (8 sections, 10 bullets)
  • Add a clear, concise description (20-200 characters ideal)
  • Replace vague words: things

actionability (60%)

  • Body contains actionable instructions (0 action items, 3 numbered)
  • No inputs defined (not always needed)
  • Include code examples or tool references in the body

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (15%)

  • Structured body describes expected behavior
  • Add examples showing how to use this skill
  • Add examples with concrete parameter values

trust (100%)

  • No executable code or suspicious patterns detected

Needs work skill with 10 suggestions for improvement (weakest: testability)

Readme

F
skill unknown Score: 46%
completeness 55% weight 20%
clarity 65% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: documentation Malice: high Capability: high

block: profile=documentation, malice=high, capability=high, categories=DESTRUCTIVE, INJECTION, SECRET

Details & Suggestions

completeness (55%)

  • Rich runbook body (6265 words, 29 sections, 17 bullets)
  • Has 9 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (65%)

  • Well-structured body (29 sections, 17 bullets)
  • Language is concrete and specific
  • Add a clear, concise description (20-200 characters ideal)

actionability (75%)

  • Body contains actionable instructions (0 action items, 6 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 6 executable code block(s): bash
  • [bash] (cd ~/.claude/skills/gstack && ./setup --team) && ~/.claude/skills/gstack/bin/gs
  • [bash] git clone --single-branch --depth 1 https://github.com/garrytan/gstack.git ~/gst
  • [bash] ~/.claude/skills/gstack/bin/gstack-uninstall
  • [bash] # 1. Stop browse daemons
  • [bash] /setup-gbrain
  • [bash] gstack-brain-init
  • Contains 6 inline command(s):
  • $ bash (cd ~/.claude/skills/gstack && ./setup --team) && ~/.claude/skills/gstack/b...
  • $ bash git clone --single-branch --depth 1 https://github.com/garrytan/gstack.git ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-uninstall
  • $ bash # 1. Stop browse daemons pkill -f "gstack.*browse" 2>/dev/null || true # 2...
  • $ bash /setup-gbrain
  • $ bash gstack-brain-init
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Destructive rm -rf on root or home directory
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)
  • [SECRET] Possible BIP39 mnemonic seed phrase

Poor skill with 11 suggestions for improvement (weakest: trust)

gstack

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: developer-toolkit Malice: medium Capability: medium

human_review: profile=developer-toolkit, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (6284 words, 38 sections, 80 bullets)
  • Has 28 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (38 sections, 80 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (2 action items, 16 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 25 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] # 1. Go to the page
  • [bash] $B goto https://yourapp.com
  • [bash] # Navigate to the feature
  • [bash] # Quick: 3 screenshots at mobile/tablet/desktop
  • [bash] $B goto https://app.example.com/upload
  • [bash] $B goto https://app.example.com/form
  • [bash] # Set up dialog handling BEFORE triggering
  • [bash] # Import cookies from your real browser (opens interactive picker)
  • [bash] $B diff https://staging.app.com https://prod.app.com
  • [bash] echo '[
  • [bash] # Element exists and is visible
  • [bash] $B click @e3 $B fill @e4 "value" $B hover @e1
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 25 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash # 1. Go to the page $B goto https://app.example.com/login # 2. See what's ...
  • $ bash $B goto https://yourapp.com $B text # read the pag...
  • $ bash # Navigate to the feature $B goto https://app.example.com/new-feature # Ta...
  • $ bash # Quick: 3 screenshots at mobile/tablet/desktop $B goto https://yourapp.com...
  • $ bash $B goto https://app.example.com/upload $B snapshot -i $B upload @e3 /path/t...
  • $ bash $B goto https://app.example.com/form $B snapshot -i # Submit empty — check...
  • $ bash # Set up dialog handling BEFORE triggering $B dialog-accept # ...
  • $ bash # Import cookies from your real browser (opens interactive picker) $B cooki...
  • $ bash $B diff https://staging.app.com https://prod.app.com
  • $ bash echo '[ ["goto","https://app.example.com"], ["snapshot","-i"], ["fill...
  • $ bash # Element exists and is visible $B is visible ".modal" # Button is enabled...
  • $ bash $B click @e3 $B fill @e4 "value" $B hover @e1 $B html @e2 ...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

Todos

D
skill unknown Score: 55%
completeness 43% weight 20%
clarity 65% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 45% weight 10%
trust 65% weight 20%
Verdict: human_review Profile: documentation Malice: medium Capability: low

human_review: profile=documentation, malice=medium, capability=low, categories=EXFILTRATION

Details & Suggestions

completeness (43%)

  • Rich runbook body (15166 words, 169 sections, 62 bullets)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add usage examples showing how to invoke this skill
  • Extract warnings into a dedicated ## Gotchas section

clarity (65%)

  • Well-structured body (169 sections, 62 bullets)
  • Language is concrete and specific
  • Add a clear, concise description (20-200 characters ideal)

actionability (75%)

  • Body contains actionable instructions (3 action items, 10 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (45%)

  • Body contains inline examples
  • Structured body describes expected behavior
  • Add examples with concrete parameter values

trust (65%)

  • HIGH: 1 suspicious pattern(s) found
  • [EXFILTRATION] Accesses SSH keys

Needs work skill with 8 suggestions for improvement (weakest: safety)

Using Gbrain With Gstack

F
skill unknown Score: 46%
completeness 55% weight 20%
clarity 65% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: human_review Profile: developer-toolkit Malice: high Capability: medium

human_review: profile=developer-toolkit, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PERSISTENCE

Details & Suggestions

completeness (55%)

  • Rich runbook body (4132 words, 34 sections, 26 bullets)
  • Has 13 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a description explaining what this skill does
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (65%)

  • Well-structured body (34 sections, 26 bullets)
  • Language is concrete and specific
  • Add a clear, concise description (20-200 characters ideal)

actionability (75%)

  • Body contains actionable instructions (1 action items, 5 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 12 executable code block(s): bash
  • [bash] /setup-gbrain
  • [bash] claude mcp add gbrain -- gbrain serve
  • [bash] /setup-gbrain --repo # re-prompt for this repo only
  • [bash] ~/.claude/skills/gstack/bin/gstack-gbrain-repo-policy list
  • [bash] /sync-gbrain # incremental: mtime fast-path, ~seconds on a clean
  • [bash] gbrain sync --source <source-id> --skip-failed
  • [bash] /setup-gbrain --switch
  • [bash] gstack-brain-init
  • [bash] /setup-gbrain --cleanup-orphans
  • [bash] /setup-gbrain --resume-provision <ref>
  • [bash] rm -rf ~/.gstack/.setup-gbrain.lock.d
  • [bash] gbrain sync --source <source-id> --skip-failed
  • Contains 12 inline command(s):
  • $ bash /setup-gbrain
  • $ bash claude mcp add gbrain -- gbrain serve
  • $ bash /setup-gbrain --repo # re-prompt for this repo only # Or directly: ~/...
  • $ bash ~/.claude/skills/gstack/bin/gstack-gbrain-repo-policy list
  • $ bash /sync-gbrain # incremental: mtime fast-path, ~seconds on a c...
  • $ bash gbrain sync --source <source-id> --skip-failed
  • $ bash /setup-gbrain --switch
  • $ bash gstack-brain-init
  • $ bash /setup-gbrain --cleanup-orphans
  • $ git worktree
  • $ bash /setup-gbrain --resume-provision <ref>
  • $ bash rm -rf ~/.gstack/.setup-gbrain.lock.d
  • CRITICAL: 3 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Destructive rm -rf on root or home directory
  • [EXFILTRATION] May leak secrets (credential in output)
  • [PERSISTENCE] Modifying shell profile — persistence via shell init

Poor skill with 9 suggestions for improvement (weakest: trust)

autoplan

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (11730 words, 59 sections, 327 bullets)
  • Has 39 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (59 sections, 327 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (19 action items, 39 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 30 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"autoplan","question_i
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] _TEL=$(~/.claude/skills/gstack/bin/gstack-config get telemetry 2>/dev/null || ec
  • [bash] _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo"
  • [bash] _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo"
  • [bash] _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo"
  • [bash] _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] COMMIT=$(git rev-parse --short HEAD 2>/dev/null)
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-design-review","ti
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-devex-review","tim
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"autoplan-voices","times
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"autoplan-voices","times
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"autoplan-voices","times
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 33 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"autoplan","quest...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat SLUG=$(~/.claude/skills...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash _TEL=$(~/.claude/skills/gstack/bin/gstack-config get telemetry 2>/dev/null ...
  • $ bash _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a gi...
  • $ bash _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a gi...
  • $ bash _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a gi...
  • $ bash _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a gi...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" TASKS_DIR="${...
  • $ bash COMMIT=$(git rev-parse --short HEAD 2>/dev/null) TIMESTAMP=$(date -u +%Y-%m...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-design-review...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-devex-review"...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"autoplan-voices","...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"autoplan-voices","...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"autoplan-voices","...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

benchmark

F
skill claude-native Score: 44%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 25% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: developer-toolkit Malice: medium Capability: medium

human_review: profile=developer-toolkit, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Has 11 steps
  • Has 26 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters
  • Too many steps (11) — consider grouping or splitting the skill

actionability (25%)

  • No inputs defined (not always needed)
  • Start each step with an action verb (Run, Check, Verify, etc.)
  • Reference specific tools or commands in steps (e.g. `git diff`, Grep)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 20 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null || echo "SLUG=unknow
  • [bash] git diff $(gh pr view --json baseRefName -q .baseRefName 2>/dev/null || gh repo
  • [bash] $B goto <page-url>
  • [bash] $B eval "JSON.stringify(performance.getEntriesByType('navigation')[0])"
  • [bash] $B eval "JSON.stringify(performance.getEntriesByType('resource').map(r => ({name
  • [bash] $B eval "JSON.stringify(performance.getEntriesByType('resource').filter(r => r.i
  • [bash] $B eval "(() => { const r = performance.getEntriesByType('resource'); return JSO
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 20 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null || echo "SLUG=u...
  • $ bash git diff $(gh pr view --json baseRefName -q .baseRefName 2>/dev/null || gh ...
  • $ bash $B goto <page-url> $B perf
  • $ bash $B eval "JSON.stringify(performance.getEntriesByType('navigation')[0])"
  • $ bash $B eval "JSON.stringify(performance.getEntriesByType('resource').map(r => (...
  • $ bash $B eval "JSON.stringify(performance.getEntriesByType('resource').filter(r =...
  • $ bash $B eval "(() => { const r = performance.getEntriesByType('resource'); retur...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Poor skill with 13 suggestions for improvement (weakest: trust)

benchmark-models

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: developer-toolkit Malice: medium Capability: medium

human_review: profile=developer-toolkit, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (3539 words, 19 sections, 59 bullets)
  • Has 16 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (19 sections, 59 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (1 action items, 5 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 15 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] BIN="$HOME/.claude/skills/gstack/bin/gstack-model-benchmark"
  • [bash] "$BIN" --prompt "unused, dry-run" --models claude,gpt,gemini --dry-run
  • [bash] [ -n "$ANTHROPIC_API_KEY" ] || grep -q 'ANTHROPIC' "$HOME/.claude/.credentials.j
  • [bash] "$BIN" <prompt-spec> --models <picked-models> [--judge] --output table
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 15 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash BIN="$HOME/.claude/skills/gstack/bin/gstack-model-benchmark" [ -x "$BIN" ] ...
  • $ bash "$BIN" --prompt "unused, dry-run" --models claude,gpt,gemini --dry-run
  • $ bash [ -n "$ANTHROPIC_API_KEY" ] || grep -q 'ANTHROPIC' "$HOME/.claude/.credenti...
  • $ bash "$BIN" <prompt-spec> --models <picked-models> [--judge] --output table
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

browse

C
skill claude-native Score: 74%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 100% weight 20%
Verdict: warn Profile: general Malice: low Capability: low

warn: no active security findings; profile=general

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (6095 words, 45 sections, 45 bullets)
  • Has 35 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (45 sections, 45 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (1 action items, 8 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (100%)

  • Contains 32 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] $B goto https://yourapp.com
  • [bash] $B goto https://app.com/login
  • [bash] $B snapshot # baseline
  • [bash] $B snapshot -i -a -o /tmp/annotated.png # labeled screenshot
  • [bash] $B snapshot -C # finds divs with cursor:pointer, onclick, tabi
  • [bash] $B is visible ".modal"
  • [bash] $B responsive /tmp/layout # mobile + tablet + desktop screenshots
  • [bash] $B upload "#file-input" /path/to/file.pdf
  • [bash] $B dialog-accept "yes" # set up handler
  • [bash] $B diff https://staging.app.com https://prod.app.com
  • [bash] # HTML file on disk → goto file:// (absolute, or cwd-relative)
  • [bash] $B viewport 480x600 --scale 2 # 2x deviceScaleFactor
  • [bash] # Generate HTML in memory, render at 2x scale, screenshot the tweet card.
  • [bash] # 1. Open a visible Chrome at the current page
  • [bash] # Headed mode — visible Chromium window. Auto-spawns Xvfb on Linux
  • [bash] $B click @e3 $B fill @e4 "value" $B hover @e1
  • [bash] $B inspect .header # full CSS cascade for selector
  • [bash] $B style .header background-color #1a1a1a # modify CSS property
  • [bash] $B cleanup --all # remove ads, cookies, sticky, social
  • Includes 1 companion script(s): build-node-server.sh
  • Contains 32 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash $B goto https://yourapp.com $B text # content load...
  • $ bash $B goto https://app.com/login $B snapshot -i # see all in...
  • $ bash $B snapshot # baseline $B click @e3 ...
  • $ bash $B snapshot -i -a -o /tmp/annotated.png # labeled screenshot $B screensho...
  • $ bash $B snapshot -C # finds divs with cursor:pointer, onclick,...
  • $ bash $B is visible ".modal" $B is enabled "#submit-btn" $B is disabled "#submit-...
  • $ bash $B responsive /tmp/layout # mobile + tablet + desktop screenshots $B...
  • $ bash $B upload "#file-input" /path/to/file.pdf $B is visible ".upload-success"
  • $ bash $B dialog-accept "yes" # set up handler $B click "#delete-button"...
  • $ bash $B diff https://staging.app.com https://prod.app.com
  • $ bash # HTML file on disk → goto file:// (absolute, or cwd-relative) $B goto file...
  • $ bash $B viewport 480x600 --scale 2 # 2x deviceScaleFactor $B load-html /tm...
  • $ bash # Generate HTML in memory, render at 2x scale, screenshot the tweet card. e...
  • $ bash # 1. Open a visible Chrome at the current page $B handoff "Stuck on CAPTCHA...
  • $ bash # Headed mode — visible Chromium window. Auto-spawns Xvfb on Linux # contai...
  • $ bash $B click @e3 $B fill @e4 "value" $B hover @e1 $B html @e2 ...
  • $ bash $B inspect .header # full CSS cascade for selector $B inspect ...
  • $ bash $B style .header background-color #1a1a1a # modify CSS property $B style ...
  • $ bash $B cleanup --all # remove ads, cookies, sticky, social $B c...
  • Executable code found — no suspicious patterns detected

Acceptable skill with 5 suggestions for improvement (weakest: safety)

canary

F
skill claude-native Score: 44%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 25% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: deployment Malice: medium Capability: medium

human_review: profile=deployment, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Has 20 steps
  • Has 29 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters
  • Too many steps (20) — consider grouping or splitting the skill

actionability (25%)

  • No inputs defined (not always needed)
  • Start each step with an action verb (Run, Check, Verify, etc.)
  • Reference specific tools or commands in steps (e.g. `git diff`, Grep)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 23 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"canary","question_id"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null || echo "SLUG=unknow
  • [bash] $B goto <page-url>
  • [bash] $B goto <url>
  • [bash] $B goto <page-url>
  • [bash] $B goto <page-url>
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 26 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"canary","questio...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null || echo "SLUG=u...
  • $ bash $B goto <page-url> $B snapshot -i -a -o ".gstack/canary-reports/baselines/<...
  • $ bash $B goto <url> $B links $B snapshot -i
  • $ bash $B goto <page-url> $B snapshot -i -a -o ".gstack/canary-reports/screenshots...
  • $ bash $B goto <page-url> $B snapshot -i -a -o ".gstack/canary-reports/screenshots...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" mkdir -p ~/.g...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Poor skill with 13 suggestions for improvement (weakest: trust)

careful

F
skill claude-native Score: 43%
completeness 80% weight 20%
clarity 70% weight 15%
actionability 50% weight 20%
safety 5% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (283 words, 3 sections, 1 bullets)
  • Has 1 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (70%)

  • Structured body (3 sections, 1 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (50%)

  • Runbook body present
  • Body references tools or includes code
  • No inputs defined (not always needed)
  • Add action-oriented instructions (bullet points starting with verbs)

safety (5%)

  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)
  • Address what happens when things go wrong (errors, failures, retries)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 1 executable code block(s): bash
  • [bash] mkdir -p ~/.gstack/analytics
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 14 inline command(s):
  • $ bash mkdir -p ~/.gstack/analytics echo '{"skill":"careful","ts":"'$(date -u +%Y-...
  • $ rm -r
  • $ rm --recursive
  • $ rm -rf /var/data
  • $ git push --force
  • $ git push -f origin main
  • $ git reset --hard
  • $ git reset --hard HEAD~3
  • $ git checkout .
  • $ git restore .
  • $ docker rm -f
  • $ docker system prune
  • $ docker system prune -a
  • $ rm -rf node_modules
  • CRITICAL: 8 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Destructive rm -rf on root or home directory
  • [DESTRUCTIVE] Destructive rm -rf on absolute path
  • [DESTRUCTIVE] Force push can destroy remote history
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Poor skill with 15 suggestions for improvement (weakest: trust)

codex

D
skill claude-native Score: 52%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 10% weight 20%
Verdict: human_review Profile: developer-toolkit Malice: medium Capability: medium

human_review: profile=developer-toolkit, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (10259 words, 43 sections, 180 bullets)
  • Has 45 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (43 sections, 180 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever, things

actionability (75%)

  • Body contains actionable instructions (5 action items, 46 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (10%)

  • Contains 32 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"codex","question_id":
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] CODEX_BIN=$(command -v codex || echo "")
  • [bash] _TEL=$(~/.claude/skills/gstack/bin/gstack-config get telemetry 2>/dev/null || ec
  • [bash] _TEL=$(~/.claude/skills/gstack/bin/gstack-config get telemetry 2>/dev/null || ec
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • [bash] TMPERR=$(mktemp "$TMP_ROOT/codex-err-XXXXXX.txt")
  • [bash] _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo"
  • [bash] _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo"
  • [bash] grep "tokens used" "$TMPERR" 2>/dev/null || echo "tokens: unknown"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"codex-review","timestam
  • [bash] rm -f "$TMPERR"
  • [bash] _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo"
  • [bash] cat .context/codex-session-id 2>/dev/null || echo "NO_SESSION"
  • [bash] TMPRESP=$(mktemp "$TMP_ROOT/codex-resp-XXXXXX.txt")
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo"
  • [bash] _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 37 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"codex","question...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash CODEX_BIN=$(command -v codex || echo "") [ -z "$CODEX_BIN" ] && echo "NOT_F...
  • $ bash _TEL=$(~/.claude/skills/gstack/bin/gstack-config get telemetry 2>/dev/null ...
  • $ bash _TEL=$(~/.claude/skills/gstack/bin/gstack-config get telemetry 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • $ git diff origin/<base> --stat 2>/dev/null | tail -1 || git diff <base> --stat 2>...
  • $ bash TMPERR=$(mktemp "$TMP_ROOT/codex-err-XXXXXX.txt")
  • $ bash _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git ...
  • $ bash _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git ...
  • $ bash grep "tokens used" "$TMPERR" 2>/dev/null || echo "tokens: unknown"
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"codex-review","tim...
  • $ bash rm -f "$TMPERR"
  • $ bash _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git ...
  • $ bash cat .context/codex-session-id 2>/dev/null || echo "NO_SESSION"
  • $ bash TMPRESP=$(mktemp "$TMP_ROOT/codex-resp-XXXXXX.txt") TMPERR=$(mktemp "$TMP_R...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat ls -t "$PLAN_ROOT"/*.md...
  • $ bash _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git ...
  • $ bash _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git ...
  • $ bash mkdir -p .context
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Printing credentials (credential logging)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 12 suggestions for improvement (weakest: trust)

open-gstack-browser

D
skill claude-native Score: 52%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 10% weight 20%
Verdict: human_review Profile: browser-automation Malice: medium Capability: medium

human_review: profile=browser-automation, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5831 words, 32 sections, 149 bullets)
  • Has 27 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (32 sections, 149 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever

actionability (75%)

  • Body contains actionable instructions (4 action items, 13 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (10%)

  • Contains 24 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"open-gstack-browser",
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] # Kill any existing browse server
  • [bash] $B connect
  • [bash] $B status
  • [bash] cat "$(git rev-parse --show-toplevel 2>/dev/null)/.gstack/browse.json" 2>/dev/nu
  • [bash] _EXT_PATH=""
  • [bash] $B goto https://news.ycombinator.com
  • [bash] $B snapshot -i
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 24 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"open-gstack-brow...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash # Kill any existing browse server if [ -f "$(git rev-parse --show-toplevel ...
  • $ bash $B connect
  • $ bash $B status
  • $ bash cat "$(git rev-parse --show-toplevel 2>/dev/null)/.gstack/browse.json" 2>/d...
  • $ bash _EXT_PATH="" _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) [ -n "$_ROO...
  • $ bash $B goto https://news.ycombinator.com
  • $ bash $B snapshot -i
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [DESTRUCTIVE] Force kill can corrupt state
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 12 suggestions for improvement (weakest: trust)

context-restore

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5055 words, 32 sections, 150 bullets)
  • Has 19 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (32 sections, 150 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (4 action items, 6 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 15 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"context-restore","que
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 15 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"context-restore"...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

context-save

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5476 words, 37 sections, 145 bullets)
  • Has 26 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (37 sections, 145 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (4 action items, 10 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 19 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"context-save","questi
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] echo "=== BRANCH ==="
  • [bash] if [ -n "$_TEL_START" ]; then
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 19 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"context-save","q...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash echo "=== BRANCH ===" git rev-parse --abbrev-ref HEAD 2>/dev/null echo "===...
  • $ bash if [ -n "$_TEL_START" ]; then START_EPOCH="$_TEL_START" elif [ -n "$PPID"...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

cso

F
skill claude-native Score: 48%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 61% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: general Malice: high Capability: high

block: profile=general, malice=high, capability=high, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE, SUSPICIOUS_URL

Details & Suggestions

completeness (80%)

  • Has description
  • Has 5 steps
  • Has 35 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Step count (5) is ideal
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever, things

actionability (61%)

  • Steps reference specific tools or commands
  • No inputs defined (not always needed)
  • Start each step with an action verb (Run, Check, Verify, etc.)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 25 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"cso","question_id":"<
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] ls package.json tsconfig.json 2>/dev/null && echo "STACK: Node/TypeScript"
  • [bash] grep -q "next" package.json 2>/dev/null && echo "FRAMEWORK: Next.js"
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] git log -p --all -S "AKIA" --diff-filter=A -- "*.env" "*.yml" "*.yaml" "*.json"
  • [bash] git ls-files '*.env' '.env.*' 2>/dev/null | grep -v '.example\|.sample\|.templat
  • [bash] for f in $(find .github/workflows -maxdepth 1 \( -name '*.yml' -o -name '*.yaml'
  • [bash] [ -f package.json ] && echo "DETECTED: npm/yarn/bun"
  • [bash] ls -la .claude/skills/ 2>/dev/null
  • [bash] mkdir -p .gstack/security-reports
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"cso","type":"TYPE","
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 28 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"cso","question_i...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash ls package.json tsconfig.json 2>/dev/null && echo "STACK: Node/TypeScript" ...
  • $ bash grep -q "next" package.json 2>/dev/null && echo "FRAMEWORK: Next.js" grep -...
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat { find .github/workflow...
  • $ bash git log -p --all -S "AKIA" --diff-filter=A -- "*.env" "*.yml" "*.yaml" "*.j...
  • $ bash git ls-files '*.env' '.env.*' 2>/dev/null | grep -v '.example\|.sample\|.te...
  • $ bash for f in $(find .github/workflows -maxdepth 1 \( -name '*.yml' -o -name '*....
  • $ git log -p --all
  • $ git log -p <base>..HEAD
  • $ npm audit
  • $ bash [ -f package.json ] && echo "DETECTED: npm/yarn/bun" [ -f Gemfile ] && echo...
  • $ bash ls -la .claude/skills/ 2>/dev/null
  • $ bash mkdir -p .gstack/security-reports
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"cso","type":"TY...
  • CRITICAL: 7 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)
  • [SUSPICIOUS_URL] Direct IP address (no DNS = suspicious)

Poor skill with 14 suggestions for improvement (weakest: trust)

design-consultation

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: general Malice: high Capability: high

block: profile=general, malice=high, capability=high, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE, SUSPICIOUS_URL

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (10203 words, 53 sections, 247 bullets)
  • Has 44 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (53 sections, 247 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (5 action items, 38 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 37 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"design-consultation",
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] ls DESIGN.md design-system.md 2>/dev/null || echo "NO_DESIGN_FILE"
  • [bash] cat README.md 2>/dev/null | head -50
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] _TASTE_PROFILE=~/.gstack/projects/$SLUG/taste-profile.json
  • [bash] $B goto "https://example-site.com"
  • [bash] command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_NOT_AV
  • [bash] TMPERR_DESIGN=$(mktemp /tmp/codex-design-XXXXXXXX)
  • [bash] cat "$TMPERR_DESIGN" && rm -f "$TMPERR_DESIGN"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"design-outside-voices",
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] $D variants --brief "<product name: [name]. Product type: [type]. Aesthetic: [di
  • [bash] $D check --image "$_DESIGN_DIR/variant-A.png" --brief "<the original brief>"
  • [bash] $D compare --images "$_DESIGN_DIR/variant-A.png,$_DESIGN_DIR/variant-B.png,$_DES
  • [bash] if [ -f "$_DESIGN_DIR/feedback.json" ]; then
  • [bash] echo '{"approved_variant":"<V>","feedback":"<FB>","date":"'$(date -u +%Y-%m-%dT%
  • [bash] PREVIEW_FILE="/tmp/design-consultation-preview-$(date +%s).html"
  • [bash] open "$PREVIEW_FILE"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"design-consultation"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 38 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"design-consultat...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash ls DESIGN.md design-system.md 2>/dev/null || echo "NO_DESIGN_FILE"
  • $ bash cat README.md 2>/dev/null | head -50 cat package.json 2>/dev/null | head -2...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat eval "$(~/.claude/skill...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) D="" [ -n "$_ROOT" ] && ...
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash _TASTE_PROFILE=~/.gstack/projects/$SLUG/taste-profile.json if [ -f "$_TASTE...
  • $ bash $B goto "https://example-site.com" $B screenshot "/tmp/design-research-site...
  • $ bash command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_N...
  • $ bash TMPERR_DESIGN=$(mktemp /tmp/codex-design-XXXXXXXX) _REPO_ROOT=$(git rev-par...
  • $ bash cat "$TMPERR_DESIGN" && rm -f "$TMPERR_DESIGN"
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"design-outside-voi...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _DESIGN_DIR="...
  • $ bash $D variants --brief "<product name: [name]. Product type: [type]. Aesthetic...
  • $ bash $D check --image "$_DESIGN_DIR/variant-A.png" --brief "<the original brief>...
  • $ bash $D compare --images "$_DESIGN_DIR/variant-A.png,$_DESIGN_DIR/variant-B.png,...
  • $ bash if [ -f "$_DESIGN_DIR/feedback.json" ]; then echo "SUBMIT_RECEIVED" cat...
  • $ curl -s -X POST http://127.0.0.1:PORT/api/reload -H 'Content-Type: application/j...
  • $ bash echo '{"approved_variant":"<V>","feedback":"<FB>","date":"'$(date -u +%Y-%m...
  • $ bash PREVIEW_FILE="/tmp/design-consultation-preview-$(date +%s).html"
  • $ bash open "$PREVIEW_FILE"
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"design-consulta...
  • CRITICAL: 8 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] Posts data to external URL
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)
  • [SUSPICIOUS_URL] Direct IP address (no DNS = suspicious)

Needs work skill with 14 suggestions for improvement (weakest: trust)

design-html

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (8553 words, 53 sections, 199 bullets)
  • Has 40 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (53 sections, 199 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things, stuff

actionability (75%)

  • Body contains actionable instructions (4 action items, 17 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 34 executable code block(s): bash, js
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"design-html","questio
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] setopt +o nomatch 2>/dev/null || true
  • [bash] setopt +o nomatch 2>/dev/null || true
  • [bash] setopt +o nomatch 2>/dev/null || true
  • [bash] setopt +o nomatch 2>/dev/null || true
  • [bash] $D prompt --image <approved-variant.png> --output json
  • [bash] [ -f package.json ] && cat package.json | grep -o '"react"\|"svelte"\|"vue"\|"@a
  • [bash] _PRETEXT_VENDOR=""
  • [bash] # Detect package manager
  • [js] import { prepare, layout } from './pretext-inline.js'
  • [js] import { prepareWithSegments, walkLineRanges } from './pretext-inline.js'
  • [js] import { prepareWithSegments, layoutNextLine } from './pretext-inline.js'
  • [js] import { prepareWithSegments, layoutWithLines } from './pretext-inline.js'
  • [bash] # Start a simple HTTP server in the output directory
  • [bash] open <path-to-finalized.html>
  • [bash] kill $_SERVER_PID 2>/dev/null || true
  • [bash] $B goto "file://<path-to-finalized.html>"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 30 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"design-html","qu...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) D="" [ -n "$_ROOT" ] && ...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • $ bash setopt +o nomatch 2>/dev/null || true _CEO=$(ls -t ~/.gstack/projects/$SLUG...
  • $ bash setopt +o nomatch 2>/dev/null || true _APPROVED=$(ls -t ~/.gstack/projects/...
  • $ bash setopt +o nomatch 2>/dev/null || true _VARIANTS=$(ls -t ~/.gstack/projects/...
  • $ bash setopt +o nomatch 2>/dev/null || true _FINALIZED=$(ls -t ~/.gstack/projects...
  • $ bash $D prompt --image <approved-variant.png> --output json
  • $ bash [ -f package.json ] && cat package.json | grep -o '"react"\|"svelte"\|"vue"...
  • $ bash _PRETEXT_VENDOR="" _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) [ -n ...
  • $ bash # Detect package manager [ -f bun.lockb ] && echo "bun add @chenglou/pretex...
  • $ bash # Start a simple HTTP server in the output directory _OUTPUT_DIR=$(dirname ...
  • $ bash open <path-to-finalized.html>
  • $ bash kill $_SERVER_PID 2>/dev/null || true
  • $ bash $B goto "file://<path-to-finalized.html>" $B screenshot /tmp/gstack-verify-...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

design-review

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: developer-toolkit Malice: high Capability: medium

block: profile=developer-toolkit, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE, SECRET

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (12770 words, 85 sections, 383 bullets)
  • Has 48 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (85 sections, 383 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things, stuff

actionability (75%)

  • Body contains actionable instructions (13 action items, 99 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 41 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"design-review","quest
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] $B status 2>/dev/null | grep -q "Mode: cdp" && echo "CDP_MODE=true" || echo "CDP
  • [bash] git status --porcelain
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] # Run the full test suite to confirm everything works
  • [bash] # Check CI provider
  • [bash] git status --porcelain
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] # Fonts in use (capped at 500 elements to avoid timeout)
  • [bash] $B goto <url>
  • [bash] $B url
  • [bash] $B snapshot -i
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_NOT_AV
  • [bash] TMPERR_DESIGN=$(mktemp /tmp/codex-design-XXXXXXXX)
  • [bash] cat "$TMPERR_DESIGN" && rm -f "$TMPERR_DESIGN"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"design-outside-voices",
  • [bash] # Search for CSS classes, component names, style files
  • [bash] $D generate --brief "<description of the page/component with the finding fixed,
  • [bash] git add <only-changed-files>
  • [bash] $B goto <affected-url>
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"design-review","type
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 42 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"design-review","...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash $B status 2>/dev/null | grep -q "Mode: cdp" && echo "CDP_MODE=true" || echo...
  • $ bash git status --porcelain
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat # Detect project runtim...
  • $ git checkout -- package.json package-lock.json
  • $ git log --since=30.days --name-only --format="" | sort | uniq -c | sort -rn | he...
  • $ bash # Run the full test suite to confirm everything works {detected test comman...
  • $ bash # Check CI provider ls -d .github/ 2>/dev/null && echo "CI:github" ls .gitl...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) D="" [ -n "$_ROOT" ] && ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" REPORT_DIR="$...
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ git diff main...HEAD --name-only
  • $ bash # Fonts in use (capped at 500 elements to avoid timeout) $B js "JSON.string...
  • $ bash $B goto <url> $B snapshot -i -a -o "$REPORT_DIR/screenshots/{page}-annotate...
  • $ bash $B url
  • $ bash $B snapshot -i $B click @e3 # perform action $B snapshot -D ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_N...
  • $ bash TMPERR_DESIGN=$(mktemp /tmp/codex-design-XXXXXXXX) _REPO_ROOT=$(git rev-par...
  • $ bash cat "$TMPERR_DESIGN" && rm -f "$TMPERR_DESIGN"
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"design-outside-voi...
  • $ bash # Search for CSS classes, component names, style files # Glob for file patt...
  • $ bash $D generate --brief "<description of the page/component with the finding fi...
  • $ bash git add <only-changed-files> git commit -m "style(design): FINDING-NNN — sh...
  • $ bash $B goto <affected-url> $B screenshot "$REPORT_DIR/screenshots/finding-NNN-a...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"design-review",...
  • CRITICAL: 7 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)
  • [SECRET] Possible BIP39 mnemonic seed phrase

Needs work skill with 13 suggestions for improvement (weakest: trust)

design-shotgun

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE, SUSPICIOUS_URL

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (7881 words, 43 sections, 159 bullets)
  • Has 35 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (43 sections, 159 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever, things, stuff

actionability (75%)

  • Body contains actionable instructions (4 action items, 38 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 28 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"design-shotgun","ques
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] cat DESIGN.md 2>/dev/null | head -80 || echo "NO_DESIGN_MD"
  • [bash] ls src/ app/ pages/ components/ 2>/dev/null | head -30
  • [bash] setopt +o nomatch 2>/dev/null || true
  • [bash] curl -s -o /dev/null -w "%{http_code}" http://localhost:3000 2>/dev/null || echo
  • [bash] _TASTE_PROFILE=~/.gstack/projects/$SLUG/taste-profile.json
  • [bash] setopt +o nomatch 2>/dev/null || true
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] $B screenshot "$_DESIGN_DIR/current.png"
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] $D compare --images "$_DESIGN_DIR/variant-A.png,$_DESIGN_DIR/variant-B.png,$_DES
  • [bash] if [ -f "$_DESIGN_DIR/feedback.json" ]; then
  • [bash] echo '{"approved_variant":"<V>","feedback":"<FB>","date":"'$(date -u +%Y-%m-%dT%
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 29 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"design-shotgun",...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) D="" [ -n "$_ROOT" ] && ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" setopt +o nom...
  • $ bash cat DESIGN.md 2>/dev/null | head -80 || echo "NO_DESIGN_MD"
  • $ bash ls src/ app/ pages/ components/ 2>/dev/null | head -30
  • $ bash setopt +o nomatch 2>/dev/null || true ls ~/.gstack/projects/$SLUG/*office-h...
  • $ bash curl -s -o /dev/null -w "%{http_code}" http://localhost:3000 2>/dev/null ||...
  • $ bash _TASTE_PROFILE=~/.gstack/projects/$SLUG/taste-profile.json if [ -f "$_TASTE...
  • $ bash setopt +o nomatch 2>/dev/null || true _TASTE=$(find ~/.gstack/projects/$SLU...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _DESIGN_DIR="...
  • $ bash $B screenshot "$_DESIGN_DIR/current.png"
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat _IMAGES=$(ls "$_DESIGN_...
  • $ bash $D compare --images "$_DESIGN_DIR/variant-A.png,$_DESIGN_DIR/variant-B.png,...
  • $ bash if [ -f "$_DESIGN_DIR/feedback.json" ]; then echo "SUBMIT_RECEIVED" cat...
  • $ curl -s -X POST http://127.0.0.1:PORT/api/reload -H 'Content-Type: application/j...
  • $ bash echo '{"approved_variant":"<V>","feedback":"<FB>","date":"'$(date -u +%Y-%m...
  • CRITICAL: 7 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] Posts data to external URL
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [SUSPICIOUS_URL] Direct IP address (no DNS = suspicious)

Needs work skill with 13 suggestions for improvement (weakest: trust)

devex-review

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: developer-toolkit Malice: medium Capability: medium

human_review: profile=developer-toolkit, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (8224 words, 55 sections, 195 bullets)
  • Has 29 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (55 sections, 195 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (11 action items, 43 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 22 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"devex-review","questi
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"devex-review","timestam
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"devex-review","type"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 25 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"devex-review","q...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" ~/.claude/ski...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"devex-review","tim...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-read
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"devex-review","...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

document-generate

F
skill claude-native Score: 40%
completeness 80% weight 20%
clarity 40% weight 15%
actionability 25% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Has 2 steps
  • Has 31 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (40%)

  • Description is too short — expand to 20-200 characters
  • Consider adding more steps (3-10 is ideal)
  • Replace vague words: things, maybe

actionability (25%)

  • No inputs defined (not always needed)
  • Start each step with an action verb (Run, Check, Verify, etc.)
  • Reference specific tools or commands in steps (e.g. `git diff`, Grep)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 18 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"document-generate","q
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] find . -type f -not -path "./.git/*" -not -path "./node_modules/*" -not -path ".
  • [bash] git commit -m "$(cat <<'EOF'
  • [bash] git push
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 24 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"document-generat...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash find . -type f -not -path "./.git/*" -not -path "./node_modules/*" -not -pa...
  • $ bash [exact command]
  • $ bash [exact command]
  • $ git add .
  • $ bash git commit -m "$(cat <<'EOF' docs: generate [scope] documentation (Diataxis...
  • $ bash git push
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Poor skill with 14 suggestions for improvement (weakest: trust)

document-release

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: deployment Malice: medium Capability: medium

human_review: profile=deployment, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (7531 words, 34 sections, 207 bullets)
  • Has 39 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (34 sections, 207 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever, things

actionability (75%)

  • Body contains actionable instructions (6 action items, 54 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 33 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"document-release","qu
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] git diff <base>...HEAD --stat
  • [bash] git log <base>..HEAD --oneline
  • [bash] git diff <base>...HEAD --name-only
  • [bash] find . -maxdepth 2 -name "*.md" -not -path "./.git/*" -not -path "./node_modules
  • [bash] git diff <base>...HEAD -- VERSION
  • [bash] git commit -m "$(cat <<'EOF'
  • [bash] git push
  • [bash] gh pr view --json body -q .body > /tmp/gstack-pr-body-$$.md
  • [bash] glab mr view -F json 2>/dev/null | python3 -c "import sys,json; print(json.load(
  • [bash] gh pr edit --body-file /tmp/gstack-pr-body-$$.md
  • [bash] glab mr update -d "$(cat <<'MRBODY'
  • [bash] rm -f /tmp/gstack-pr-body-$$.md
  • [bash] V=$(cat VERSION 2>/dev/null | tr -d '[:space:]')
  • [bash] CURRENT_TITLE=$(gh pr view --json title -q .title 2>/dev/null || true)
  • [bash] CURRENT_TITLE=$(glab mr view -F json 2>/dev/null | jq -r .title 2>/dev/null || t
  • [bash] NEW_TITLE=$(~/.claude/skills/gstack/bin/gstack-pr-title-rewrite.sh "$V" "$CURREN
  • [bash] gh pr edit --title "$NEW_TITLE"
  • [bash] glab mr update -t "$NEW_TITLE"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 37 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"document-release...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash git diff <base>...HEAD --stat
  • $ bash git log <base>..HEAD --oneline
  • $ bash git diff <base>...HEAD --name-only
  • $ bash find . -maxdepth 2 -name "*.md" -not -path "./.git/*" -not -path "./node_mo...
  • $ bash git diff <base>...HEAD -- VERSION
  • $ git add .
  • $ bash git commit -m "$(cat <<'EOF' docs: update project documentation for vX.Y.Z....
  • $ bash git push
  • $ bash gh pr view --json body -q .body > /tmp/gstack-pr-body-$$.md
  • $ bash glab mr view -F json 2>/dev/null | python3 -c "import sys,json; print(json....
  • $ bash gh pr edit --body-file /tmp/gstack-pr-body-$$.md
  • $ bash glab mr update -d "$(cat <<'MRBODY' <paste the file contents here> MRBODY )...
  • $ bash rm -f /tmp/gstack-pr-body-$$.md
  • $ bash V=$(cat VERSION 2>/dev/null | tr -d '[:space:]')
  • $ bash CURRENT_TITLE=$(gh pr view --json title -q .title 2>/dev/null || true)
  • $ bash CURRENT_TITLE=$(glab mr view -F json 2>/dev/null | jq -r .title 2>/dev/null...
  • $ bash NEW_TITLE=$(~/.claude/skills/gstack/bin/gstack-pr-title-rewrite.sh "$V" "$C...
  • $ bash gh pr edit --title "$NEW_TITLE"
  • $ bash glab mr update -t "$NEW_TITLE"
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

freeze

D
skill claude-native Score: 59%
completeness 85% weight 20%
clarity 60% weight 15%
actionability 65% weight 20%
safety 75% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (85%)

  • Has description
  • Has 2 steps
  • Has 3 code block(s) in body (inline examples)
  • Has 4 gotchas
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability

clarity (60%)

  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters
  • Consider adding more steps (3-10 is ideal)

actionability (65%)

  • Steps start with action verbs
  • No inputs defined (not always needed)
  • Reference specific tools or commands in steps (e.g. `git diff`, Grep)

safety (75%)

  • Has 4 gotchas
  • Gotchas are specific and detailed
  • Address what happens when things go wrong (errors, failures, retries)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 3 executable code block(s): bash
  • [bash] mkdir -p ~/.gstack/analytics
  • [bash] FREEZE_DIR=$(cd "<user-provided-path>" 2>/dev/null && pwd)
  • [bash] FREEZE_DIR="${FREEZE_DIR%/}/"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 3 inline command(s):
  • $ bash mkdir -p ~/.gstack/analytics echo '{"skill":"freeze","ts":"'$(date -u +%Y-%...
  • $ bash FREEZE_DIR=$(cd "<user-provided-path>" 2>/dev/null && pwd) echo "$FREEZE_DI...
  • $ bash FREEZE_DIR="${FREEZE_DIR%/}/" eval "$(~/.claude/skills/gstack/bin/gstack-pa...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

gstack-upgrade

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: developer-toolkit Malice: medium Capability: medium

human_review: profile=developer-toolkit, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (1326 words, 11 sections, 5 bullets)
  • Has 17 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (11 sections, 5 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever

actionability (75%)

  • Body contains actionable instructions (0 action items, 3 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 16 executable code block(s): bash
  • [bash] _AUTO=""
  • [bash] ~/.claude/skills/gstack/bin/gstack-config set auto_upgrade true
  • [bash] _SNOOZE_FILE="$HOME/.gstack/update-snoozed"
  • [bash] ~/.claude/skills/gstack/bin/gstack-config set update_check false
  • [bash] if [ -d "$HOME/.claude/skills/gstack/.git" ]; then
  • [bash] OLD_VERSION=$(cat "$INSTALL_DIR/VERSION" 2>/dev/null || echo "unknown")
  • [bash] cd "$INSTALL_DIR"
  • [bash] PARENT=$(dirname "$INSTALL_DIR")
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] cd "$_ROOT"
  • [bash] mv "$LOCAL_GSTACK" "$LOCAL_GSTACK.bak"
  • [bash] rm -rf "$LOCAL_GSTACK"
  • [bash] MIGRATIONS_DIR="$INSTALL_DIR/gstack-upgrade/migrations"
  • [bash] mkdir -p ~/.gstack
  • [bash] ~/.claude/skills/gstack/bin/gstack-update-check --force 2>/dev/null || \
  • [bash] PRIMARY_VER=$(cat "$INSTALL_DIR/VERSION" 2>/dev/null || echo "unknown")
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 17 inline command(s):
  • $ bash _AUTO="" [ "${GSTACK_AUTO_UPGRADE:-}" = "1" ] && _AUTO="true" [ -z "$_AUTO"...
  • $ bash ~/.claude/skills/gstack/bin/gstack-config set auto_upgrade true
  • $ bash _SNOOZE_FILE="$HOME/.gstack/update-snoozed" _REMOTE_VER="{new}" _CUR_LEVEL=...
  • $ bash ~/.claude/skills/gstack/bin/gstack-config set update_check false
  • $ bash if [ -d "$HOME/.claude/skills/gstack/.git" ]; then INSTALL_TYPE="global-g...
  • $ bash OLD_VERSION=$(cat "$INSTALL_DIR/VERSION" 2>/dev/null || echo "unknown")
  • $ bash cd "$INSTALL_DIR" STASH_OUTPUT=$(git stash 2>&1) git fetch origin git reset...
  • $ git stash pop
  • $ bash PARENT=$(dirname "$INSTALL_DIR") TMP_DIR=$(mktemp -d) git clone --depth 1 h...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) LOCAL_GSTACK="" if [ -n ...
  • $ bash cd "$_ROOT" git rm -r --cached .claude/skills/gstack/ 2>/dev/null || true i...
  • $ bash mv "$LOCAL_GSTACK" "$LOCAL_GSTACK.bak" cp -Rf "$INSTALL_DIR" "$LOCAL_GSTACK...
  • $ bash rm -rf "$LOCAL_GSTACK" mv "$LOCAL_GSTACK.bak" "$LOCAL_GSTACK"
  • $ bash MIGRATIONS_DIR="$INSTALL_DIR/gstack-upgrade/migrations" if [ -d "$MIGRATION...
  • $ bash mkdir -p ~/.gstack echo "$OLD_VERSION" > ~/.gstack/just-upgraded-from rm -f...
  • $ bash ~/.claude/skills/gstack/bin/gstack-update-check --force 2>/dev/null || \ .c...
  • $ bash PRIMARY_VER=$(cat "$INSTALL_DIR/VERSION" 2>/dev/null || echo "unknown") LOC...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

guard

F
skill claude-native Score: 48%
completeness 80% weight 20%
clarity 70% weight 15%
actionability 60% weight 20%
safety 5% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (282 words, 2 sections, 6 bullets)
  • Has 3 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (70%)

  • Structured body (2 sections, 6 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (60%)

  • Body has some actionable content
  • Body references tools or includes code
  • No inputs defined (not always needed)
  • Add more action-oriented bullet points or numbered steps

safety (5%)

  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)
  • Address what happens when things go wrong (errors, failures, retries)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 3 executable code block(s): bash
  • [bash] mkdir -p ~/.gstack/analytics
  • [bash] FREEZE_DIR=$(cd "<user-provided-path>" 2>/dev/null && pwd)
  • [bash] FREEZE_DIR="${FREEZE_DIR%/}/"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 3 inline command(s):
  • $ bash mkdir -p ~/.gstack/analytics echo '{"skill":"guard","ts":"'$(date -u +%Y-%m...
  • $ bash FREEZE_DIR=$(cd "<user-provided-path>" 2>/dev/null && pwd) echo "$FREEZE_DI...
  • $ bash FREEZE_DIR="${FREEZE_DIR%/}/" eval "$(~/.claude/skills/gstack/bin/gstack-pa...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Poor skill with 12 suggestions for improvement (weakest: safety)

health

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (6085 words, 31 sections, 163 bullets)
  • Has 30 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (31 sections, 163 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (5 action items, 24 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 18 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"health","question_id"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] # Type checker
  • [bash] # Example for each tool — run each independently
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 18 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"health","questio...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash # Type checker [ -f tsconfig.json ] && echo "TYPECHECK: tsc --noEmit" # Li...
  • $ bash # Example for each tool — run each independently START=$(date +%s) tsc --no...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

investigate

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: general Malice: high Capability: high

block: profile=general, malice=high, capability=high, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (6058 words, 33 sections, 152 bullets)
  • Has 27 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (33 sections, 152 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things, maybe

actionability (75%)

  • Body contains actionable instructions (4 action items, 19 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 21 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"investigate","questio
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git log --oneline -20 -- <affected-files>
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-search --query "<your-keyword>" --l
  • [bash] [ -x "${CLAUDE_SKILL_DIR}/../freeze/bin/check-freeze.sh" ] && echo "FREEZE_AVAIL
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"investigate","type":
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"investigate","type":
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 21 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"investigate","qu...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git log --oneline -20 -- <affected-files>
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-search --query "<your-keyword>...
  • $ bash [ -x "${CLAUDE_SKILL_DIR}/../freeze/bin/check-freeze.sh" ] && echo "FREEZE_...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-paths)" STATE_DIR="$GSTACK_STATE...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"investigate","t...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"investigate","t...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Needs work skill with 12 suggestions for improvement (weakest: trust)

ios-clean

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (4974 words, 30 sections, 147 bullets)
  • Has 18 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (30 sections, 147 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (4 action items, 24 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 15 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-clean","question_
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 16 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-clean","ques...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ git restore
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

ios-design-review

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: developer-toolkit Malice: medium Capability: medium

human_review: profile=developer-toolkit, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5032 words, 29 sections, 141 bullets)
  • Has 18 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (29 sections, 141 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (4 action items, 20 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 15 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-design-review","q
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 15 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-design-revie...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

ios-fix

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (4930 words, 31 sections, 141 bullets)
  • Has 18 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (31 sections, 141 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (4 action items, 22 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 15 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-fix","question_id
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 15 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-fix","questi...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

ios-qa

C
skill claude-native Score: 68%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 70% weight 20%
Verdict: warn Profile: developer-toolkit Malice: low Capability: medium

warn: profile=developer-toolkit, malice=low, capability=medium, categories=SUSPICIOUS_URL

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5689 words, 34 sections, 153 bullets)
  • Has 21 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (34 sections, 153 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (4 action items, 23 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (70%)

  • Contains 16 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-qa","question_id"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] SESSION="$HOME/.gstack/ios-qa-session.json"
  • Includes 2 companion script(s): gen-accessors.test.ts, gen-accessors.ts
  • Contains 16 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-qa","questio...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash SESSION="$HOME/.gstack/ios-qa-session.json" if [ -f "$SESSION" ] && [ "$COL...
  • HIGH: 1 suspicious pattern(s) found
  • [SUSPICIOUS_URL] Direct IP address (no DNS = suspicious)

Acceptable skill with 6 suggestions for improvement (weakest: safety)

ios-sync

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (4908 words, 29 sections, 141 bullets)
  • Has 19 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (29 sections, 141 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (4 action items, 20 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 16 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-sync","question_i
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] swift run --package-path "$GSTACK_HOME/ios-qa/scripts/gen-accessors-tool" \
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 16 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ios-sync","quest...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash swift run --package-path "$GSTACK_HOME/ios-qa/scripts/gen-accessors-tool" \...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

land-and-deploy

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: deployment Malice: high Capability: medium

block: profile=deployment, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (12222 words, 63 sections, 235 bullets)
  • Has 70 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (63 sections, 235 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (7 action items, 40 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 62 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"land-and-deploy","que
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] gh auth status
  • [bash] gh pr view --json number,state,title,url,mergeStateStatus,mergeable,baseRefName,
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] # Check for persisted deploy config in CLAUDE.md
  • [bash] # Test gh auth (already passed in Step 1, but confirm)
  • [bash] grep -i "staging" CLAUDE.md 2>/dev/null | head -3
  • [bash] for f in $(find .github/workflows -maxdepth 1 \( -name '*.yml' -o -name '*.yaml'
  • [bash] gh pr checks --json name,targetUrl 2>/dev/null | head -20
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read 2>/dev/null
  • [bash] mkdir -p ~/.gstack/projects/$SLUG
  • [bash] gh pr checks --json name,state,status,conclusion
  • [bash] gh pr view --json mergeable -q .mergeable
  • [bash] gh pr checks --watch --fail-fast
  • [bash] BRANCH_VERSION=$(git show HEAD:VERSION 2>/dev/null | tr -d '\r\n[:space:]' || ec
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read 2>/dev/null
  • [bash] git log --oneline STORED_COMMIT..HEAD
  • [bash] cat ~/.claude/skills/gstack/review/checklist.md 2>/dev/null || echo "Checklist n
  • [bash] bun test 2>&1 | tail -10
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] gh pr view --json body -q .body
  • [bash] git log --oneline $(gh pr view --json baseRefName -q .baseRefName 2>/dev/null ||
  • [bash] git log --oneline --all-match --grep="docs:" $(gh pr view --json baseRefName -q
  • [bash] git diff --name-only $(gh pr view --json baseRefName -q .baseRefName 2>/dev/null
  • [bash] gh pr merge --auto --delete-branch
  • [bash] gh pr merge --squash --delete-branch
  • [bash] gh pr view --json state -q .state
  • [bash] gh run list --branch <base> --limit 5 --json name,status,workflowName,headSha
  • [bash] # Check for persisted deploy config in CLAUDE.md
  • [bash] eval $(~/.claude/skills/gstack/bin/gstack-diff-scope $(gh pr view --json baseRef
  • [bash] gh run list --branch <base> --limit 5 --json name,status,conclusion,headSha,work
  • [bash] gh run list --branch <base> --limit 10 --json databaseId,headSha,status,conclusi
  • [bash] gh run view <run-id> --json status,conclusion
  • [bash] fly status --app {app} 2>/dev/null
  • [bash] curl -sf {production-url} -o /dev/null -w "%{http_code}" 2>/dev/null
  • [bash] heroku releases --app {app} -n 1 2>/dev/null
  • [bash] $B goto <url>
  • [bash] $B console --errors
  • [bash] $B perf
  • [bash] $B text
  • [bash] $B snapshot -i -a -o ".gstack/deploy-reports/post-deploy.png"
  • [bash] git fetch origin <base>
  • [bash] mkdir -p .gstack/deploy-reports
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 63 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"land-and-deploy"...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash gh auth status
  • $ bash gh pr view --json number,state,title,url,mergeStateStatus,mergeable,baseRef...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" if [ ! -f ~/....
  • $ bash # Check for persisted deploy config in CLAUDE.md DEPLOY_CONFIG=$(grep -A 20...
  • $ bash # Test gh auth (already passed in Step 1, but confirm) gh auth status 2>&1 ...
  • $ bash grep -i "staging" CLAUDE.md 2>/dev/null | head -3
  • $ bash for f in $(find .github/workflows -maxdepth 1 \( -name '*.yml' -o -name '*....
  • $ bash gh pr checks --json name,targetUrl 2>/dev/null | head -20
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-read 2>/dev/null
  • $ bash mkdir -p ~/.gstack/projects/$SLUG CURRENT_HASH=$(sed -n '/## Deploy Configu...
  • $ bash gh pr checks --json name,state,status,conclusion
  • $ bash gh pr view --json mergeable -q .mergeable
  • $ bash gh pr checks --watch --fail-fast
  • $ bash BRANCH_VERSION=$(git show HEAD:VERSION 2>/dev/null | tr -d '\r\n[:space:]' ...
  • $ bash git log --oneline STORED_COMMIT..HEAD
  • $ bash cat ~/.claude/skills/gstack/review/checklist.md 2>/dev/null || echo "Checkl...
  • $ bash bun test 2>&1 | tail -10
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat ls -t ~/.gstack-dev/eva...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat ls -t ~/.gstack-dev/eva...
  • $ bash gh pr view --json body -q .body
  • $ bash git log --oneline $(gh pr view --json baseRefName -q .baseRefName 2>/dev/nu...
  • $ bash git log --oneline --all-match --grep="docs:" $(gh pr view --json baseRefNam...
  • $ bash git diff --name-only $(gh pr view --json baseRefName -q .baseRefName 2>/dev...
  • $ bash gh pr merge --auto --delete-branch
  • $ bash gh pr merge --squash --delete-branch
  • $ bash gh pr view --json state -q .state
  • $ bash gh run list --branch <base> --limit 5 --json name,status,workflowName,headS...
  • $ bash eval $(~/.claude/skills/gstack/bin/gstack-diff-scope $(gh pr view --json ba...
  • $ bash gh run list --branch <base> --limit 5 --json name,status,conclusion,headSha...
  • $ bash gh run list --branch <base> --limit 10 --json databaseId,headSha,status,con...
  • $ bash gh run view <run-id> --json status,conclusion
  • $ bash fly status --app {app} 2>/dev/null
  • $ bash curl -sf {production-url} -o /dev/null -w "%{http_code}" 2>/dev/null
  • $ bash heroku releases --app {app} -n 1 2>/dev/null
  • $ bash $B goto <url>
  • $ bash $B console --errors
  • $ bash $B perf
  • $ bash $B text
  • $ bash $B snapshot -i -a -o ".gstack/deploy-reports/post-deploy.png"
  • $ bash git fetch origin <base> git checkout <base> git revert <merge-commit-sha> -...
  • $ bash mkdir -p .gstack/deploy-reports
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" mkdir -p ~/.g...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Needs work skill with 12 suggestions for improvement (weakest: trust)

landing-report

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5177 words, 31 sections, 146 bullets)
  • Has 23 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (31 sections, 146 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (4 action items, 9 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 18 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"landing-report","ques
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] BASE_BRANCH=$(gh pr view --json baseRefName -q .baseRefName 2>/dev/null || \
  • [bash] CURRENT_VERSION=$(cat VERSION 2>/dev/null | tr -d '[:space:]' || echo "0.0.0.0")
  • [bash] for LEVEL in micro patch minor major; do
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 18 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"landing-report",...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash BASE_BRANCH=$(gh pr view --json baseRefName -q .baseRefName 2>/dev/null || ...
  • $ bash CURRENT_VERSION=$(cat VERSION 2>/dev/null | tr -d '[:space:]' || echo "0.0....
  • $ bash for LEVEL in micro patch minor major; do bun run bin/gstack-next-version ...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

learn

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5050 words, 34 sections, 151 bullets)
  • Has 24 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (34 sections, 151 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (4 action items, 13 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 20 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"learn","question_id":
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"learn","type":"TYPE"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 20 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"learn","question...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" ~/.claude/ski...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" ~/.claude/ski...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" ~/.claude/ski...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" ~/.claude/ski...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" eval "$(~/.cl...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"learn","type":"...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

make-pdf

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (3518 words, 22 sections, 56 bullets)
  • Has 20 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (22 sections, 56 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (2 action items, 5 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 17 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] $P generate letter.md # writes /tmp/letter.pdf
  • [bash] $P generate --cover --toc --author "Garry Tan" --title "On Horizons" \
  • [bash] $P generate --watermark DRAFT memo.md draft.pdf
  • [bash] $P preview essay.md
  • [bash] $P generate --no-confidential memo.md memo.pdf
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 17 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) P="" [ -n "$MAKE_PDF_BIN...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash $P generate letter.md # writes /tmp/letter.pdf $P generate ...
  • $ bash $P generate --cover --toc --author "Garry Tan" --title "On Horizons" \ es...
  • $ bash $P generate --watermark DRAFT memo.md draft.pdf
  • $ bash $P preview essay.md
  • $ bash $P generate --no-confidential memo.md memo.pdf
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

office-hours

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: general Malice: high Capability: high

block: profile=general, malice=high, capability=high, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (15346 words, 98 sections, 268 bullets)
  • Has 53 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (98 sections, 268 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things, maybe

actionability (75%)

  • Body contains actionable instructions (6 action items, 93 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 44 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"office-hours","questi
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_NOT_AV
  • [bash] CODEX_PROMPT_FILE=$(mktemp /tmp/gstack-codex-oh-XXXXXXXX.txt)
  • [bash] TMPERR_OH=$(mktemp /tmp/codex-oh-err-XXXXXXXX)
  • [bash] cat "$TMPERR_OH"
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] $D variants --brief "<assembled brief>" --count 3 --output-dir "$_DESIGN_DIR/"
  • [bash] $D compare --images "$_DESIGN_DIR/variant-A.png,$_DESIGN_DIR/variant-B.png,$_DES
  • [bash] echo '{"approved_variant":"<VARIANT>","feedback":"<FEEDBACK>","date":"'$(date -u
  • [bash] SKETCH_FILE="/tmp/gstack-sketch-$(date +%s).html"
  • [bash] $B goto "file://$SKETCH_FILE"
  • [bash] command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_NOT_AV
  • [bash] TMPERR_SKETCH=$(mktemp /tmp/codex-sketch-XXXXXXXX)
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] mkdir -p ~/.gstack/analytics
  • [bash] PROFILE=$(~/.claude/skills/gstack/bin/gstack-builder-profile 2>/dev/null) || PRO
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • [bash] mkdir -p ~/.gstack/analytics
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"office-hours","type"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 46 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"office-hours","q...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • $ git log --oneline -30
  • $ git diff origin/main --stat 2>/dev/null
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat ls -t ~/.gstack/p...
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat grep -li "<keyword1>\|<...
  • $ bash command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_N...
  • $ bash CODEX_PROMPT_FILE=$(mktemp /tmp/gstack-codex-oh-XXXXXXXX.txt)
  • $ bash TMPERR_OH=$(mktemp /tmp/codex-oh-err-XXXXXXXX) _REPO_ROOT=$(git rev-parse -...
  • $ bash cat "$TMPERR_OH" rm -f "$TMPERR_OH" "$CODEX_PROMPT_FILE"
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) D="" [ -n "$_ROOT" ] && ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _DESIGN_DIR="...
  • $ bash $D variants --brief "<assembled brief>" --count 3 --output-dir "$_DESIGN_DI...
  • $ bash $D compare --images "$_DESIGN_DIR/variant-A.png,$_DESIGN_DIR/variant-B.png,...
  • $ curl -X POST http://localhost:PORT/api/reload -H 'Content-Type: application/json...
  • $ bash echo '{"approved_variant":"<VARIANT>","feedback":"<FEEDBACK>","date":"'$(da...
  • $ bash SKETCH_FILE="/tmp/gstack-sketch-$(date +%s).html"
  • $ bash $B goto "file://$SKETCH_FILE" $B screenshot /tmp/gstack-sketch.png
  • $ bash TMPERR_SKETCH=$(mktemp /tmp/codex-sketch-XXXXXXXX) _REPO_ROOT=$(git rev-par...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-paths)" mkdir -p "$GSTACK_STATE_...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-paths)" echo '{"date":"TIMESTAMP...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat PRIOR=$(ls -t ~/.gstack...
  • $ bash mkdir -p ~/.gstack/analytics echo '{"skill":"office-hours","ts":"'$(date -u...
  • $ bash PROFILE=$(~/.claude/skills/gstack/bin/gstack-builder-profile 2>/dev/null) |...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-paths)" open "$GSTACK_STATE_ROOT...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-paths)" echo '{"date":"'"$(date ...
  • $ bash mkdir -p ~/.gstack/analytics echo '{"skill":"office-hours","event":"resourc...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"office-hours","...
  • CRITICAL: 7 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] Posts data to external URL
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Needs work skill with 13 suggestions for improvement (weakest: trust)

open-gstack-browser

D
skill claude-native Score: 52%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 10% weight 20%
Verdict: human_review Profile: browser-automation Malice: medium Capability: medium

human_review: profile=browser-automation, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5831 words, 32 sections, 149 bullets)
  • Has 27 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (32 sections, 149 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever

actionability (75%)

  • Body contains actionable instructions (4 action items, 13 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (10%)

  • Contains 24 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"open-gstack-browser",
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] # Kill any existing browse server
  • [bash] $B connect
  • [bash] $B status
  • [bash] cat "$(git rev-parse --show-toplevel 2>/dev/null)/.gstack/browse.json" 2>/dev/nu
  • [bash] _EXT_PATH=""
  • [bash] $B goto https://news.ycombinator.com
  • [bash] $B snapshot -i
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 24 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"open-gstack-brow...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash # Kill any existing browse server if [ -f "$(git rev-parse --show-toplevel ...
  • $ bash $B connect
  • $ bash $B status
  • $ bash cat "$(git rev-parse --show-toplevel 2>/dev/null)/.gstack/browse.json" 2>/d...
  • $ bash _EXT_PATH="" _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) [ -n "$_ROO...
  • $ bash $B goto https://news.ycombinator.com
  • $ bash $B snapshot -i
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [DESTRUCTIVE] Force kill can corrupt state
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 12 suggestions for improvement (weakest: trust)

pair-agent

D
skill claude-native Score: 57%
completeness 80% weight 20%
clarity 80% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5915 words, 40 sections, 159 bullets)
  • Has 30 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (80%)

  • Well-structured body (40 sections, 159 bullets)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters

actionability (75%)

  • Body contains actionable instructions (7 action items, 16 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 27 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"pair-agent","question
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] $B status 2>/dev/null
  • [bash] $B goto about:blank
  • [bash] $B pair-agent --local TARGET_HOST
  • [bash] which ngrok 2>/dev/null && echo "NGROK_INSTALLED" || echo "NGROK_NOT_INSTALLED"
  • [bash] $B pair-agent --client TARGET_HOST
  • [bash] $B pair-agent --admin --client TARGET_HOST
  • [bash] ngrok config add-authtoken THEIR_TOKEN
  • [bash] $B status
  • [bash] $B tunnel revoke AGENT_NAME
  • [bash] # This invalidates ALL scoped tokens immediately
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 27 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"pair-agent","que...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash $B status 2>/dev/null
  • $ bash $B goto about:blank
  • $ bash $B pair-agent --local TARGET_HOST
  • $ bash which ngrok 2>/dev/null && echo "NGROK_INSTALLED" || echo "NGROK_NOT_INSTAL...
  • $ bash $B pair-agent --client TARGET_HOST
  • $ bash $B pair-agent --admin --client TARGET_HOST
  • $ bash ngrok config add-authtoken THEIR_TOKEN
  • $ bash $B status
  • $ bash $B tunnel revoke AGENT_NAME
  • $ bash # This invalidates ALL scoped tokens immediately $B tunnel rotate
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 10 suggestions for improvement (weakest: trust)

plan-ceo-review

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: developer-toolkit Malice: high Capability: medium

block: profile=developer-toolkit, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (17968 words, 100 sections, 380 bullets)
  • Has 51 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (100 sections, 380 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (12 action items, 88 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 32 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-ceo-review","que
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] mkdir -p ~/.gstack/projects/$SLUG/ceo-plans/archive
  • [bash] mkdir -p ~/.gstack/analytics
  • [bash] command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_NOT_AV
  • [bash] TMPERR_PV=$(mktemp /tmp/codex-planreview-XXXXXXXX)
  • [bash] cat "$TMPERR_PV"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"codex-plan-review","tim
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-ceo-review","times
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"plan-ceo-review","ty
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 35 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-ceo-review"...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ git log --oneline -30 # Recent history git diff <base> ...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat SLUG=$(~/.claude/skills...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat HANDOFF=$(ls -t ~/.gsta...
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash mkdir -p ~/.gstack/projects/$SLUG/ceo-plans/archive # For each stale plan: ...
  • $ bash mkdir -p ~/.gstack/analytics echo '{"skill":"plan-ceo-review","ts":"'$(date...
  • $ bash command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_N...
  • $ bash TMPERR_PV=$(mktemp /tmp/codex-planreview-XXXXXXXX) _REPO_ROOT=$(git rev-par...
  • $ bash cat "$TMPERR_PV"
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"codex-plan-review"...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" TASKS_DIR="${...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat eval "$(~/.claude/skill...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-ceo-review","...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-read
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"plan-ceo-review...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Needs work skill with 12 suggestions for improvement (weakest: trust)

plan-design-review

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: developer-toolkit Malice: high Capability: medium

block: profile=developer-toolkit, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE, SUSPICIOUS_URL

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (14326 words, 86 sections, 286 bullets)
  • Has 46 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (86 sections, 286 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things, stuff

actionability (75%)

  • Body contains actionable instructions (7 action items, 105 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 34 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-design-review","
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] git log --oneline -15
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] $D variants --brief "<description assembled from plan + DESIGN.md constraints>"
  • [bash] $D check --image "$_DESIGN_DIR/variant-A.png" --brief "<the original brief>"
  • [bash] $D compare --images "$_DESIGN_DIR/variant-A.png,$_DESIGN_DIR/variant-B.png,$_DES
  • [bash] if [ -f "$_DESIGN_DIR/feedback.json" ]; then
  • [bash] echo '{"approved_variant":"<V>","feedback":"<FB>","date":"'$(date -u +%Y-%m-%dT%
  • [bash] command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_NOT_AV
  • [bash] TMPERR_DESIGN=$(mktemp /tmp/codex-design-XXXXXXXX)
  • [bash] cat "$TMPERR_DESIGN" && rm -f "$TMPERR_DESIGN"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"design-outside-voices",
  • [bash] $D generate --brief "<description of what 10/10 looks like for this dimension>"
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-design-review","ti
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"plan-design-review",
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 38 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-design-revi...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash git log --oneline -15 git diff <base> --stat
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) D="" [ -n "$_ROOT" ] && ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _DESIGN_DIR="...
  • $ bash $D variants --brief "<description assembled from plan + DESIGN.md constrain...
  • $ bash $D check --image "$_DESIGN_DIR/variant-A.png" --brief "<the original brief>...
  • $ bash $D compare --images "$_DESIGN_DIR/variant-A.png,$_DESIGN_DIR/variant-B.png,...
  • $ bash if [ -f "$_DESIGN_DIR/feedback.json" ]; then echo "SUBMIT_RECEIVED" cat...
  • $ curl -s -X POST http://127.0.0.1:PORT/api/reload -H 'Content-Type: application/j...
  • $ bash echo '{"approved_variant":"<V>","feedback":"<FB>","date":"'$(date -u +%Y-%m...
  • $ bash command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_N...
  • $ bash TMPERR_DESIGN=$(mktemp /tmp/codex-design-XXXXXXXX) _REPO_ROOT=$(git rev-par...
  • $ bash cat "$TMPERR_DESIGN" && rm -f "$TMPERR_DESIGN"
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"design-outside-voi...
  • $ bash $D generate --brief "<description of what 10/10 looks like for this dimensi...
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" TASKS_DIR="${...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-design-review...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-read
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"plan-design-rev...
  • CRITICAL: 8 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] Posts data to external URL
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)
  • [SUSPICIOUS_URL] Direct IP address (no DNS = suspicious)

Needs work skill with 14 suggestions for improvement (weakest: trust)

plan-devex-review

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: developer-toolkit Malice: high Capability: medium

block: profile=developer-toolkit, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (14296 words, 86 sections, 304 bullets)
  • Has 43 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (86 sections, 304 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever, things

actionability (75%)

  • Body contains actionable instructions (7 action items, 61 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 28 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-devex-review","q
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] git log --oneline -15
  • [bash] setopt +o nomatch 2>/dev/null || true
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_NOT_AV
  • [bash] TMPERR_PV=$(mktemp /tmp/codex-planreview-XXXXXXXX)
  • [bash] cat "$TMPERR_PV"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"codex-plan-review","tim
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"plan-devex-review","
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 32 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-devex-revie...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash git log --oneline -15 git diff $(git merge-base HEAD main 2>/dev/null || ec...
  • $ bash setopt +o nomatch 2>/dev/null || true SLUG=$(~/.claude/skills/gstack/browse...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat SLUG=$(~/.claude/skills...
  • $ docker run hello-world
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" ~/.claude/ski...
  • $ bash command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_N...
  • $ bash TMPERR_PV=$(mktemp /tmp/codex-planreview-XXXXXXXX) _REPO_ROOT=$(git rev-par...
  • $ bash cat "$TMPERR_PV"
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"codex-plan-review"...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" TASKS_DIR="${...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-read
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"plan-devex-revi...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Needs work skill with 12 suggestions for improvement (weakest: trust)

plan-eng-review

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: developer-toolkit Malice: high Capability: medium

block: profile=developer-toolkit, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (13687 words, 75 sections, 295 bullets)
  • Has 37 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (75 sections, 295 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (8 action items, 51 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 28 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-eng-review","que
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_NOT_AV
  • [bash] TMPERR_PV=$(mktemp /tmp/codex-planreview-XXXXXXXX)
  • [bash] cat "$TMPERR_PV"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"codex-plan-review","tim
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-eng-review","times
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"plan-eng-review","ty
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 27 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-eng-review"...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat SLUG=$(~/.claude/skills...
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat # Detect project runtim...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_N...
  • $ bash TMPERR_PV=$(mktemp /tmp/codex-planreview-XXXXXXXX) _REPO_ROOT=$(git rev-par...
  • $ bash cat "$TMPERR_PV"
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"codex-plan-review"...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" TASKS_DIR="${...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"plan-eng-review","...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-read
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"plan-eng-review...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Needs work skill with 12 suggestions for improvement (weakest: trust)

plan-tune

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (6325 words, 31 sections, 151 bullets)
  • Has 26 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (31 sections, 151 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (5 action items, 31 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 23 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-tune","question_
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _QT=$(~/.claude/skills/gstack/bin/gstack-config get question_tuning 2>/dev/null
  • [bash] ~/.claude/skills/gstack/bin/gstack-config set question_tuning true
  • [bash] # Ensure profile exists
  • [bash] ~/.claude/skills/gstack/bin/gstack-developer-profile --profile
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-developer-profile --gap
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --stats
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 23 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"plan-tune","ques...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _QT=$(~/.claude/skills/gstack/bin/gstack-config get question_tuning 2>/d...
  • $ bash ~/.claude/skills/gstack/bin/gstack-config set question_tuning true
  • $ bash # Ensure profile exists ~/.claude/skills/gstack/bin/gstack-developer-...
  • $ bash ~/.claude/skills/gstack/bin/gstack-developer-profile --profile
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" eval "$(~/.cl...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"questi...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-paths)" _PROFILE="$GSTACK_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-developer-profile --gap
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --stats eval "$(~/.c...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

qa

F
skill claude-native Score: 40%
completeness 80% weight 20%
clarity 40% weight 15%
actionability 38% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: developer-toolkit Malice: high Capability: medium

block: profile=developer-toolkit, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Has 32 steps
  • Has 50 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (40%)

  • Description is too short — expand to 20-200 characters
  • Too many steps (32) — consider grouping or splitting the skill
  • Replace vague words: things

actionability (38%)

  • No inputs defined (not always needed)
  • Start each step with an action verb (Run, Check, Verify, etc.)
  • Reference specific tools or commands in steps (e.g. `git diff`, Grep)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 43 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"qa","question_id":"<i
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] $B status 2>/dev/null | grep -q "Mode: cdp" && echo "CDP_MODE=true" || echo "CDP
  • [bash] git status --porcelain
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] # Run the full test suite to confirm everything works
  • [bash] # Check CI provider
  • [bash] git status --porcelain
  • [bash] mkdir -p .gstack/qa-reports/screenshots
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] git diff main...HEAD --name-only
  • [bash] $B goto http://localhost:3000 2>/dev/null && echo "Found app on :3000" || \
  • [bash] $B goto <login-url>
  • [bash] $B cookie-import cookies.json
  • [bash] $B goto <target-url>
  • [bash] $B goto <page-url>
  • [bash] $B viewport 375x812
  • [bash] $B screenshot "$REPORT_DIR/screenshots/issue-001-step-1.png"
  • [bash] $B snapshot -i -a -o "$REPORT_DIR/screenshots/issue-002.png"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-search --query "<your-keyword>" --l
  • [bash] # Grep for error messages, component names, route definitions
  • [bash] git add <only-changed-files>
  • [bash] $B goto <affected-url>
  • [bash] {detected test command} {new-test-file}
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"qa","type":"TYPE","k
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 47 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"qa","question_id...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash $B status 2>/dev/null | grep -q "Mode: cdp" && echo "CDP_MODE=true" || echo...
  • $ bash git status --porcelain
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat # Detect project runtim...
  • $ git checkout -- package.json package-lock.json
  • $ git log --since=30.days --name-only --format="" | sort | uniq -c | sort -rn | he...
  • $ bash # Run the full test suite to confirm everything works {detected test comman...
  • $ bash # Check CI provider ls -d .github/ 2>/dev/null && echo "CI:github" ls .gitl...
  • $ bash mkdir -p .gstack/qa-reports/screenshots
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat eval "$(~/.claude...
  • $ bash git diff main...HEAD --name-only git log main..HEAD --oneline
  • $ bash $B goto http://localhost:3000 2>/dev/null && echo "Found app on :3000" |...
  • $ bash $B goto <login-url> $B snapshot -i # find the login form...
  • $ bash $B cookie-import cookies.json $B goto <target-url>
  • $ bash $B goto <target-url> $B snapshot -i -a -o "$REPORT_DIR/screenshots/initial....
  • $ bash $B goto <page-url> $B snapshot -i -a -o "$REPORT_DIR/screenshots/page-name....
  • $ bash $B viewport 375x812 $B screenshot "$REPORT_DIR/screenshots/page-mobil...
  • $ bash $B screenshot "$REPORT_DIR/screenshots/issue-001-step-1.png" $B click @e5 $...
  • $ bash $B snapshot -i -a -o "$REPORT_DIR/screenshots/issue-002.png"
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-search --query "<your-keyword>...
  • $ bash # Grep for error messages, component names, route definitions # Glob for fi...
  • $ bash git add <only-changed-files> git commit -m "fix(qa): ISSUE-NNN — short desc...
  • $ bash $B goto <affected-url> $B screenshot "$REPORT_DIR/screenshots/issue-NNN-aft...
  • $ bash {detected test command} {new-test-file}
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"qa","type":"TYP...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Poor skill with 15 suggestions for improvement (weakest: trust)

qa-only

F
skill claude-native Score: 40%
completeness 80% weight 20%
clarity 40% weight 15%
actionability 38% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: developer-toolkit Malice: high Capability: medium

block: profile=developer-toolkit, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Has 32 steps
  • Has 36 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (40%)

  • Description is too short — expand to 20-200 characters
  • Too many steps (32) — consider grouping or splitting the skill
  • Replace vague words: things

actionability (38%)

  • No inputs defined (not always needed)
  • Start each step with an action verb (Run, Check, Verify, etc.)
  • Reference specific tools or commands in steps (e.g. `git diff`, Grep)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 31 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"qa-only","question_id
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] REPORT_DIR=".gstack/qa-reports"
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] git diff main...HEAD --name-only
  • [bash] $B goto http://localhost:3000 2>/dev/null && echo "Found app on :3000" || \
  • [bash] $B goto <login-url>
  • [bash] $B cookie-import cookies.json
  • [bash] $B goto <target-url>
  • [bash] $B goto <page-url>
  • [bash] $B viewport 375x812
  • [bash] $B screenshot "$REPORT_DIR/screenshots/issue-001-step-1.png"
  • [bash] $B snapshot -i -a -o "$REPORT_DIR/screenshots/issue-002.png"
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"qa-only","type":"TYP
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 31 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"qa-only","questi...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash REPORT_DIR=".gstack/qa-reports" mkdir -p "$REPORT_DIR/screenshots"
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat eval "$(~/.claude...
  • $ bash git diff main...HEAD --name-only git log main..HEAD --oneline
  • $ bash $B goto http://localhost:3000 2>/dev/null && echo "Found app on :3000" |...
  • $ bash $B goto <login-url> $B snapshot -i # find the login form...
  • $ bash $B cookie-import cookies.json $B goto <target-url>
  • $ bash $B goto <target-url> $B snapshot -i -a -o "$REPORT_DIR/screenshots/initial....
  • $ bash $B goto <page-url> $B snapshot -i -a -o "$REPORT_DIR/screenshots/page-name....
  • $ bash $B viewport 375x812 $B screenshot "$REPORT_DIR/screenshots/page-mobil...
  • $ bash $B screenshot "$REPORT_DIR/screenshots/issue-001-step-1.png" $B click @e5 $...
  • $ bash $B snapshot -i -a -o "$REPORT_DIR/screenshots/issue-002.png"
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"qa-only","type"...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Poor skill with 15 suggestions for improvement (weakest: trust)

retro

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: general Malice: high Capability: high

block: profile=general, malice=high, capability=high, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (10525 words, 81 sections, 271 bullets)
  • Has 54 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (81 sections, 271 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (14 action items, 24 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 33 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"retro","question_id":
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] [ -f ~/.gstack/retro-context.md ] && echo "RETRO_CONTEXT_FOUND" || echo "NO_RETR
  • [bash] git fetch origin <default> --quiet
  • [bash] # 1. All commits in window with timestamps, subject, hash, AUTHOR, files changed
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"retro","type":"TYPE"
  • [bash] # Team streak: all unique commit dates (local time) — no hard cutoff
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] mkdir -p .context/retros
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] DISCOVER_BIN=""
  • [bash] $DISCOVER_BIN --since "<window>" --format json 2>/tmp/gstack-discover-stderr
  • [bash] git -C <path> fetch origin --quiet 2>/dev/null
  • [bash] # Commits with stats
  • [bash] git -C <path> log origin/$DEFAULT --since="365 days ago" --format="%ad" --date=f
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] mkdir -p ~/.gstack/retros
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 37 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"retro","question...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash [ -f ~/.gstack/retro-context.md ] && echo "RETRO_CONTEXT_FOUND" || echo "NO...
  • $ bash git fetch origin <default> --quiet # Identify who is running the retro git ...
  • $ bash # 1. All commits in window with timestamps, subject, hash, AUTHOR, files ch...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"retro","type":"...
  • $ bash # Team streak: all unique commit dates (local time) — no hard cutoff git lo...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat ls -t .context/retros/*...
  • $ bash mkdir -p .context/retros
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat # Count existing retros...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat eval "$(~/.claude/skill...
  • $ bash DISCOVER_BIN="" [ -x ~/.claude/skills/gstack/bin/gstack-global-discover ] &...
  • $ bash $DISCOVER_BIN --since "<window>" --format json 2>/tmp/gstack-discover-stder...
  • $ bash git -C <path> fetch origin --quiet 2>/dev/null
  • $ git rev-parse --abbrev-ref HEAD
  • $ bash # Commits with stats git -C <path> log origin/$DEFAULT --since="<start_date...
  • $ bash git -C <path> log origin/$DEFAULT --since="365 days ago" --format="%ad" --d...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat ls -t ~/.gstack/retros/...
  • $ bash mkdir -p ~/.gstack/retros
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat today=$(date +%Y-%m-%d)...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Needs work skill with 12 suggestions for improvement (weakest: trust)

review

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: developer-toolkit Malice: high Capability: medium

block: profile=developer-toolkit, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (12078 words, 75 sections, 230 bullets)
  • Has 45 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (75 sections, 230 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (10 action items, 58 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 35 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"review","question_id"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{
  • [bash] git fetch origin <base> --quiet
  • [bash] DIFF_BASE=$(git merge-base origin/<base> HEAD)
  • [bash] BRANCH_VERSION=$(git show HEAD:VERSION 2>/dev/null | tr -d '\r\n[:space:]' || ec
  • [bash] bun run slop:diff origin/<base> 2>/dev/null || true
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] source <(~/.claude/skills/gstack/bin/gstack-diff-scope <base> 2>/dev/null) || tr
  • [bash] ~/.claude/skills/gstack/bin/gstack-specialist-stats 2>/dev/null || true
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-search --type pitfall --query "{spe
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read
  • [bash] git diff --name-only <prior-review-commit> HEAD
  • [bash] DIFF_BASE=$(git merge-base origin/<base> HEAD)
  • [bash] TMPERR_ADV=$(mktemp /tmp/codex-adv-XXXXXXXX)
  • [bash] cat "$TMPERR_ADV"
  • [bash] TMPERR=$(mktemp /tmp/codex-review-XXXXXXXX)
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"adversarial-review","ti
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"review","timestamp":"TI
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"review","type":"TYPE
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 46 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"review","questio...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ git branch --show-current
  • $ git fetch origin <base> --quiet && DIFF_BASE=$(git merge-base origin/<base> HEAD...
  • $ git log origin/<base>..HEAD --oneline
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat BRANCH=$(git branch --s...
  • $ git diff
  • $ git diff <base>...HEAD
  • $ npm run validate-wiki -- <path>
  • $ git diff origin/<base>...HEAD
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{ "type": "pitfall", ...
  • $ bash git fetch origin <base> --quiet
  • $ bash DIFF_BASE=$(git merge-base origin/<base> HEAD) git diff "$DIFF_BASE"
  • $ bash BRANCH_VERSION=$(git show HEAD:VERSION 2>/dev/null | tr -d '\r\n[:space:]' ...
  • $ bash bun run slop:diff origin/<base> 2>/dev/null || true
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash source <(~/.claude/skills/gstack/bin/gstack-diff-scope <base> 2>/dev/null) ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-specialist-stats 2>/dev/null || true
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-search --type pitfall --query ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-read
  • $ bash git diff --name-only <prior-review-commit> HEAD
  • $ bash DIFF_BASE=$(git merge-base origin/<base> HEAD) DIFF_INS=$(git diff "$DIFF_B...
  • $ bash TMPERR_ADV=$(mktemp /tmp/codex-adv-XXXXXXXX) _REPO_ROOT=$(git rev-parse --s...
  • $ bash cat "$TMPERR_ADV"
  • $ bash TMPERR=$(mktemp /tmp/codex-review-XXXXXXXX) _REPO_ROOT=$(git rev-parse --sh...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"adversarial-review...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"review","timestamp...
  • $ git rev-parse --short HEAD
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"review","type":...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Needs work skill with 12 suggestions for improvement (weakest: trust)

scrape

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: browser-automation Malice: medium Capability: medium

human_review: profile=browser-automation, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (5472 words, 33 sections, 155 bullets)
  • Has 21 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (33 sections, 155 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever, things

actionability (75%)

  • Body contains actionable instructions (5 action items, 13 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 18 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"scrape","question_id"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] $B skill list
  • [bash] $B skill run <name> [--arg key=value ...]
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"scrape","type":"TYPE
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 18 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"scrape","questio...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash $B skill list
  • $ bash $B skill run <name> [--arg key=value ...]
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"scrape","type":...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

setup-browser-cookies

D
skill claude-native Score: 64%
completeness 85% weight 20%
clarity 60% weight 15%
actionability 68% weight 20%
safety 100% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: browser-automation Malice: medium Capability: medium

human_review: profile=browser-automation, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (85%)

  • Has description
  • Has 12 steps
  • Has 18 code block(s) in body (inline examples)
  • Has 5 gotchas
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability

clarity (60%)

  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters
  • Too many steps (12) — consider grouping or splitting the skill

actionability (68%)

  • Steps reference specific tools or commands
  • No inputs defined (not always needed)
  • Start each step with an action verb (Run, Check, Verify, etc.)

safety (100%)

  • Has 5 gotchas
  • Gotchas are specific and detailed
  • Mentions error handling concepts

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 17 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] $B status 2>/dev/null | grep -q "Mode: cdp" && echo "CDP_MODE=true" || echo "CDP
  • [bash] _ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
  • [bash] if ! command -v bun >/dev/null 2>&1; then
  • [bash] $B cookie-import-browser
  • [bash] $B cookie-import-browser comet --domain github.com
  • [bash] $B cookies
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 19 inline command(s):
  • $ git rm -r .claude/skills/gstack/
  • $ git add .claude/ .gitignore CLAUDE.md && git commit -m "chore: migrate gstack fr...
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash $B status 2>/dev/null | grep -q "Mode: cdp" && echo "CDP_MODE=true" || echo...
  • $ bash _ROOT=$(git rev-parse --show-toplevel 2>/dev/null) B="" [ -n "$_ROOT" ] && ...
  • $ bash if ! command -v bun >/dev/null 2>&1; then BUN_VERSION="1.3.10" ...
  • $ bash $B cookie-import-browser
  • $ bash $B cookie-import-browser comet --domain github.com
  • $ bash $B cookies
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 9 suggestions for improvement (weakest: trust)

setup-deploy

F
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 57% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: human_review Profile: deployment Malice: medium Capability: medium

human_review: profile=deployment, malice=medium, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Has 30 steps
  • Has 23 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters
  • Too many steps (30) — consider grouping or splitting the skill

actionability (57%)

  • Steps reference specific tools or commands
  • No inputs defined (not always needed)
  • Start each step with an action verb (Run, Check, Verify, etc.)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 18 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"setup-deploy","questi
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] grep -A 20 "## Deploy Configuration" CLAUDE.md 2>/dev/null || echo "NO_CONFIG"
  • [bash] # Platform config files
  • [bash] curl -sf "{health-check-url}" -o /dev/null -w "%{http_code}" 2>/dev/null || echo
  • [bash] {deploy-status-command} 2>/dev/null | head -5 || echo "COMMAND_FAILED"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 18 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"setup-deploy","q...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash grep -A 20 "## Deploy Configuration" CLAUDE.md 2>/dev/null || echo "NO_CONF...
  • $ bash # Platform config files [ -f fly.toml ] && echo "PLATFORM:fly" && cat fly.t...
  • $ bash curl -sf "{health-check-url}" -o /dev/null -w "%{http_code}" 2>/dev/null ||...
  • $ bash {deploy-status-command} 2>/dev/null | head -5 || echo "COMMAND_FAILED"
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Poor skill with 12 suggestions for improvement (weakest: trust)

setup-gbrain

D
skill claude-native Score: 52%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 5% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (10181 words, 59 sections, 195 bullets)
  • Has 53 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (59 sections, 195 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever

actionability (75%)

  • Body contains actionable instructions (4 action items, 10 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (5%)

  • Contains 43 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"setup-gbrain","questi
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] ~/.claude/skills/gstack/bin/gstack-gbrain-detect
  • [bash] BACKUP="$HOME/.gbrain/config.json.gstack-bak-$(date +%s)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-gbrain-install
  • [bash] . ~/.claude/skills/gstack/bin/gstack-gbrain-lib.sh
  • [bash] printf '%s' "$GBRAIN_POOLER_URL" | ~/.claude/skills/gstack/bin/gstack-gbrain-sup
  • [bash] GBRAIN_DATABASE_URL="$GBRAIN_POOLER_URL" gbrain init --non-interactive --json
  • [bash] . ~/.claude/skills/gstack/bin/gstack-gbrain-lib.sh
  • [bash] orgs=$(~/.claude/skills/gstack/bin/gstack-gbrain-supabase-provision list-orgs --
  • [bash] export DB_PASS=$(openssl rand -base64 24)
  • [bash] trap 'echo ""; echo "gstack-gbrain: interrupted. In-flight ref: $INFLIGHT_REF";
  • [bash] result=$(~/.claude/skills/gstack/bin/gstack-gbrain-supabase-provision \
  • [bash] gbrain init --pglite --json
  • [bash] . ~/.claude/skills/gstack/bin/gstack-gbrain-lib.sh
  • [bash] verify_json=$(GBRAIN_MCP_TOKEN="$GBRAIN_MCP_TOKEN" \
  • [bash] ~/.claude/skills/gstack/bin/gstack-gbrain-install || exit $?
  • [bash] # Going PGLite → Supabase, collect URL first (Path 1 flow), then:
  • [bash] doctor=$(gbrain doctor --json)
  • [bash] claude mcp remove gbrain -s user 2>/dev/null || true
  • [bash] GBRAIN_BIN=$(command -v gbrain)
  • [bash] current_tier=$(~/.claude/skills/gstack/bin/gstack-gbrain-repo-policy get)
  • [bash] ~/.claude/skills/gstack/bin/gstack-gbrain-repo-policy set "$REMOTE" "$TIER"
  • [bash] URL_FORM=${URL_FORM_SUPPORTED:-false}
  • [bash] GBRAIN_URL=$(python3 -c "
  • [bash] ~/.claude/skills/gstack/bin/gstack-memory-ingest --probe
  • [bash] ~/.claude/skills/gstack/bin/gstack-config set transcript_ingest_mode <choice>
  • [bash] SLUG="setup-gbrain-smoke-test-$(date +%s)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-gbrain-detect 2>/dev/null || true
  • [bash] # List user's Supabase projects (user has to pipe this through their own
  • [bash] curl -s -X DELETE -H "Authorization: Bearer $SUPABASE_ACCESS_TOKEN" \
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 44 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"setup-gbrain","q...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash ~/.claude/skills/gstack/bin/gstack-gbrain-detect
  • $ bash BACKUP="$HOME/.gbrain/config.json.gstack-bak-$(date +%s)" mv "$HOME/.gbrain...
  • $ bash ~/.claude/skills/gstack/bin/gstack-gbrain-install
  • $ bash . ~/.claude/skills/gstack/bin/gstack-gbrain-lib.sh read_secret_to_env GBRAI...
  • $ bash printf '%s' "$GBRAIN_POOLER_URL" | ~/.claude/skills/gstack/bin/gstack-gbrai...
  • $ bash GBRAIN_DATABASE_URL="$GBRAIN_POOLER_URL" gbrain init --non-interactive --js...
  • $ bash . ~/.claude/skills/gstack/bin/gstack-gbrain-lib.sh read_secret_to_env SUPAB...
  • $ bash orgs=$(~/.claude/skills/gstack/bin/gstack-gbrain-supabase-provision list-or...
  • $ bash export DB_PASS=$(openssl rand -base64 24)
  • $ bash trap 'echo ""; echo "gstack-gbrain: interrupted. In-flight ref: $INFLIGHT_R...
  • $ bash result=$(~/.claude/skills/gstack/bin/gstack-gbrain-supabase-provision \ c...
  • $ bash gbrain init --pglite --json
  • $ bash . ~/.claude/skills/gstack/bin/gstack-gbrain-lib.sh read_secret_to_env GBRAI...
  • $ bash verify_json=$(GBRAIN_MCP_TOKEN="$GBRAIN_MCP_TOKEN" \ ~/.claude/skills/gst...
  • $ bash ~/.claude/skills/gstack/bin/gstack-gbrain-install || exit $? # At this poin...
  • $ bash # Going PGLite → Supabase, collect URL first (Path 1 flow), then: timeout 1...
  • $ bash doctor=$(gbrain doctor --json) status=$(echo "$doctor" | jq -r .status)
  • $ bash claude mcp remove gbrain -s user 2>/dev/null || true claude mcp remove gbra...
  • $ bash GBRAIN_BIN=$(command -v gbrain) [ -z "$GBRAIN_BIN" ] && GBRAIN_BIN="$HOME/....
  • $ bash current_tier=$(~/.claude/skills/gstack/bin/gstack-gbrain-repo-policy get)
  • $ bash ~/.claude/skills/gstack/bin/gstack-gbrain-repo-policy set "$REMOTE" "$TIE...
  • $ bash URL_FORM=${URL_FORM_SUPPORTED:-false} ~/.claude/skills/gstack/bin/gstack-ar...
  • $ git worktree
  • $ bash GBRAIN_URL=$(python3 -c " import json, os, sys try: c = json.load(open(...
  • $ bash ~/.claude/skills/gstack/bin/gstack-memory-ingest --probe
  • $ bash ~/.claude/skills/gstack/bin/gstack-config set transcript_ingest_mode <choic...
  • $ bash SLUG="setup-gbrain-smoke-test-$(date +%s)" echo "Set up on $(date). Smoke t...
  • $ bash ~/.claude/skills/gstack/bin/gstack-gbrain-detect 2>/dev/null || true ~/.cla...
  • $ bash # List user's Supabase projects (user has to pipe this through their own # ...
  • $ bash curl -s -X DELETE -H "Authorization: Bearer $SUPABASE_ACCESS_TOKEN" \ htt...
  • CRITICAL: 7 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Destructive rm -rf on root or home directory
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 13 suggestions for improvement (weakest: trust)

ship

D
skill claude-native Score: 50%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 0% weight 20%
Verdict: block Profile: deployment Malice: high Capability: medium

block: profile=deployment, malice=high, capability=medium, categories=DESTRUCTIVE, EXFILTRATION, INJECTION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (22387 words, 119 sections, 381 bullets)
  • Has 91 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (119 sections, 381 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (21 action items, 121 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (0%)

  • Contains 77 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ship","question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] git remote get-url origin 2>/dev/null
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read
  • [bash] git diff origin/<base> --name-only | grep -E '(cmd/.*/main\.go|bin/|Cargo\.toml|
  • [bash] ls .github/workflows/ 2>/dev/null | grep -iE 'release|publish|dist'
  • [bash] git fetch origin <base> && git merge origin/<base> --no-edit
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] # Run the full test suite to confirm everything works
  • [bash] # Check CI provider
  • [bash] git status --porcelain
  • [bash] bin/test-lane 2>&1 | tee /tmp/ship_tests.txt &
  • [bash] git diff origin/<base>...HEAD --name-only
  • [bash] # Who last touched the failing test?
  • [bash] gh issue create \
  • [bash] glab issue create \
  • [bash] git diff origin/<base> --name-only
  • [bash] grep -l "changed_file_basename" test/evals/*_eval_runner.rb
  • [bash] EVAL_JUDGE_TIER=full EVAL_VERBOSE=1 bin/test-lane --eval test/evals/<suite>_eval
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] # Count test files before any generation
  • [bash] # Count test files after generation
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] setopt +o nomatch 2>/dev/null || true # zsh compat
  • [bash] curl -s -o /dev/null -w '%{http_code}' http://localhost:3000 2>/dev/null || \
  • [bash] cat ${CLAUDE_SKILL_DIR}/../qa-only/SKILL.md
  • [bash] _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_learni
  • [bash] source <(~/.claude/skills/gstack/bin/gstack-diff-scope <base> 2>/dev/null)
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"design-review-lite","ti
  • [bash] command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_NOT_AV
  • [bash] TMPERR_DRL=$(mktemp /tmp/codex-drl-XXXXXXXX)
  • [bash] cat "$TMPERR_DRL" && rm -f "$TMPERR_DRL"
  • [bash] source <(~/.claude/skills/gstack/bin/gstack-diff-scope <base> 2>/dev/null) || tr
  • [bash] ~/.claude/skills/gstack/bin/gstack-specialist-stats 2>/dev/null || true
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-search --type pitfall --query "{spe
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-read
  • [bash] git diff --name-only <prior-review-commit> HEAD
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"review","timestamp":"TI
  • [bash] DIFF_BASE=$(git merge-base origin/<base> HEAD)
  • [bash] TMPERR_ADV=$(mktemp /tmp/codex-adv-XXXXXXXX)
  • [bash] cat "$TMPERR_ADV"
  • [bash] TMPERR=$(mktemp /tmp/codex-review-XXXXXXXX)
  • [bash] ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"adversarial-review","ti
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"ship","type":"TYPE",
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-search --query "<your-keyword>" --l
  • [bash] if ! git rev-parse --verify origin/<base> >/dev/null 2>&1; then
  • [bash] QUEUE_JSON=$(bun run bin/gstack-next-version \
  • [bash] if ! printf '%s' "$NEW_VERSION" | grep -qE '^[0-9]+\.[0-9]+\.[0-9]+\.[0-9]+$'; t
  • [bash] REPAIR_VERSION=$(cat VERSION | tr -d '\r\n[:space:]')
  • [bash] git log <base>..HEAD --oneline
  • [bash] git diff <base>...HEAD
  • [bash] WIP_COUNT=$(git log <base>..HEAD --oneline --grep="^WIP:" 2>/dev/null | wc -l |
  • [bash] # Export [gstack-context] blocks from all WIP commits on this branch.
  • [bash] # Interactive rebase with automated WIP squashing.
  • [bash] # Branch contains only WIP commits. Reset-soft is safe here because there's
  • [bash] git commit -m "$(cat <<'EOF'
  • [bash] git fetch origin <branch-name> 2>/dev/null
  • [bash] git push -u origin <branch-name>
  • [bash] gh pr view --json url,number,state -q 'if .state == "OPEN" then "PR #\(.number):
  • [bash] glab mr view -F json 2>/dev/null | jq -r 'if .state == "opened" then "MR_EXISTS"
  • [bash] # PR title MUST start with v$NEW_VERSION — enforced on every run, no exceptions.
  • [bash] # MR title MUST start with v$NEW_VERSION — enforced on every run, no exceptions.
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~/.gst
  • [bash] echo '{"skill":"ship","timestamp":"'"$(date -u +%Y-%m-%dT%H:%M:%SZ)"'","coverage
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 92 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"ship","question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash git remote get-url origin 2>/dev/null
  • $ git log
  • $ git fetch
  • $ git merge
  • $ git status
  • $ git diff <base>...HEAD --stat
  • $ git log <base>..HEAD --oneline
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-read
  • $ bash git diff origin/<base> --name-only | grep -E '(cmd/.*/main\.go|bin/|Carg...
  • $ bash ls .github/workflows/ 2>/dev/null | grep -iE 'release|publish|dist' g...
  • $ bash git fetch origin <base> && git merge origin/<base> --no-edit
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat # Detect project runtim...
  • $ git checkout -- package.json package-lock.json
  • $ git log --since=30.days --name-only --format="" | sort | uniq -c | sort -rn | he...
  • $ bash # Run the full test suite to confirm everything works {detected test comman...
  • $ bash # Check CI provider ls -d .github/ 2>/dev/null && echo "CI:github" ls .gitl...
  • $ bash git status --porcelain
  • $ bash bin/test-lane 2>&1 | tee /tmp/ship_tests.txt & npm run test 2>&1 | tee /tmp...
  • $ bash git diff origin/<base>...HEAD --name-only
  • $ git commit -m "fix: pre-existing test failure in <test-file>"
  • $ bash # Who last touched the failing test? git log --format="%an (%ae)" -1 --...
  • $ bash gh issue create \ --title "Pre-existing test failure: <test-name>...
  • $ bash glab issue create \ -t "Pre-existing test failure: <test-name>" \...
  • $ bash git diff origin/<base> --name-only
  • $ bash grep -l "changed_file_basename" test/evals/*_eval_runner.rb
  • $ bash EVAL_JUDGE_TIER=full EVAL_VERBOSE=1 bin/test-lane --eval test/evals/<suite>...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat # Detect project runtim...
  • $ bash # Count test files before any generation find . -name '*.test.*' -o -name '...
  • $ bash # Count test files after generation find . -name '*.test.*' -o -name '*.spe...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash setopt +o nomatch 2>/dev/null || true # zsh compat BRANCH=$(git branch --s...
  • $ git diff
  • $ git diff <base>...HEAD
  • $ npm run validate-wiki -- <path>
  • $ git diff origin/<base>...HEAD
  • $ git log origin/<base>..HEAD --oneline
  • $ bash curl -s -o /dev/null -w '%{http_code}' http://localhost:3000 2>/dev/null ||...
  • $ bash cat ${CLAUDE_SKILL_DIR}/../qa-only/SKILL.md
  • $ bash _CROSS_PROJ=$(~/.claude/skills/gstack/bin/gstack-config get cross_project_l...
  • $ bash source <(~/.claude/skills/gstack/bin/gstack-diff-scope <base> 2>/dev/null)
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"design-review-lite...
  • $ bash command -v codex >/dev/null 2>&1 && echo "CODEX_AVAILABLE" || echo "CODEX_N...
  • $ bash TMPERR_DRL=$(mktemp /tmp/codex-drl-XXXXXXXX) _REPO_ROOT=$(git rev-parse --s...
  • $ bash cat "$TMPERR_DRL" && rm -f "$TMPERR_DRL"
  • $ bash source <(~/.claude/skills/gstack/bin/gstack-diff-scope <base> 2>/dev/null) ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-specialist-stats 2>/dev/null || true
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-search --type pitfall --query ...
  • $ bash git diff --name-only <prior-review-commit> HEAD
  • $ git add <fixed-files> && git commit -m "fix: pre-landing review fixes"
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"review","timestamp...
  • $ bash DIFF_BASE=$(git merge-base origin/<base> HEAD) DIFF_INS=$(git diff "$DIFF_B...
  • $ bash TMPERR_ADV=$(mktemp /tmp/codex-adv-XXXXXXXX) _REPO_ROOT=$(git rev-parse --s...
  • $ bash cat "$TMPERR_ADV"
  • $ bash TMPERR=$(mktemp /tmp/codex-review-XXXXXXXX) _REPO_ROOT=$(git rev-parse --sh...
  • $ bash ~/.claude/skills/gstack/bin/gstack-review-log '{"skill":"adversarial-review...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"ship","type":"T...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-search --query "<your-keyword>...
  • $ bash if ! git rev-parse --verify origin/<base> >/dev/null 2>&1; then echo "ERR...
  • $ git diff origin/<base>...HEAD --stat | tail -1
  • $ bash QUEUE_JSON=$(bun run bin/gstack-next-version \ --base <base> \ ...
  • $ bash if ! printf '%s' "$NEW_VERSION" | grep -qE '^[0-9]+\.[0-9]+\.[0-9]+\.[0-9]+...
  • $ bash REPAIR_VERSION=$(cat VERSION | tr -d '\r\n[:space:]') if ! printf '%s' "$RE...
  • $ bash git log <base>..HEAD --oneline
  • $ bash git diff <base>...HEAD
  • $ bash WIP_COUNT=$(git log <base>..HEAD --oneline --grep="^WIP:" 2>/dev/null | wc ...
  • $ bash # Export [gstack-context] blocks from all WIP commits on this branch. # Thi...
  • $ bash # Interactive rebase with automated WIP squashing. # Mark every WIP commit ...
  • $ bash # Branch contains only WIP commits. Reset-soft is safe here because there's...
  • $ bash git commit -m "$(cat <<'EOF' chore: bump version and changelog (vX.Y.Z.W) ...
  • $ bash git fetch origin <branch-name> 2>/dev/null LOCAL=$(git rev-parse HEAD) REMO...
  • $ bash git push -u origin <branch-name>
  • $ bash gh pr view --json url,number,state -q 'if .state == "OPEN" then "PR #\(.num...
  • $ bash glab mr view -F json 2>/dev/null | jq -r 'if .state == "opened" then "MR_EX...
  • $ bash # PR title MUST start with v$NEW_VERSION — enforced on every run, no except...
  • $ bash # MR title MUST start with v$NEW_VERSION — enforced on every run, no except...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" && mkdir -p ~...
  • $ bash echo '{"skill":"ship","timestamp":"'"$(date -u +%Y-%m-%dT%H:%M:%SZ)"'","cov...
  • CRITICAL: 6 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges
  • [INJECTION] Conditional logic based on first run (rug-pull pattern)

Needs work skill with 12 suggestions for improvement (weakest: trust)

skillify

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (6839 words, 40 sections, 161 bullets)
  • Has 33 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (40 sections, 161 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: things

actionability (75%)

  • Body contains actionable instructions (5 action items, 10 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 20 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"skillify","question_i
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --arg b
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] $B goto "<TARGET_URL>"
  • [bash] $B skill test "<name>" --dir "<stagedDir>"
  • [bash] ( cd "<stagedDir>" && bun test script.test.ts )
  • [bash] $B skill list | grep <name>
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"skillify","type":"TY
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 20 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"skillify","quest...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash jq -n --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" --arg skill "SKILL_NAME" --...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash $B goto "<TARGET_URL>" $B html > /tmp/skillify-fixture-$$.html
  • $ bash $B skill test "<name>" --dir "<stagedDir>"
  • $ bash ( cd "<stagedDir>" && bun test script.test.ts )
  • $ bash $B skill list | grep <name> $B skill run <name> # should match the JSON ...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"skillify","type...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

sync-gbrain

D
skill claude-native Score: 54%
completeness 80% weight 20%
clarity 60% weight 15%
actionability 75% weight 20%
safety 30% weight 15%
testability 60% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (80%)

  • Has description
  • Rich runbook body (6217 words, 33 sections, 159 bullets)
  • Has 24 code block(s) in body (inline examples)
  • Body contains warning/caveat language
  • Runbook style (no formal inputs)
  • Add a dedicated ## Examples section for discoverability
  • Extract warnings into a dedicated ## Gotchas section

clarity (60%)

  • Well-structured body (33 sections, 159 bullets)
  • Description is too short — expand to 20-200 characters
  • Replace vague words: whatever

actionability (75%)

  • Body contains actionable instructions (4 action items, 6 numbered)
  • Body references tools or includes code
  • No inputs defined (not always needed)

safety (30%)

  • Mentions error handling concepts
  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)

testability (60%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Structured body describes expected behavior

trust (15%)

  • Contains 18 executable code block(s): bash
  • [bash] _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .claude/sk
  • [bash] rm -f ~/.gstack/.writing-style-prompt-pending
  • [bash] open https://garryslist.org/posts/boil-the-ocean
  • [bash] touch ~/.gstack/.telemetry-prompted
  • [bash] touch ~/.gstack/.proactive-prompted
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null || tru
  • [bash] _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}"
  • [bash] # Chosen mode: full | artifacts-only | off
  • [bash] "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null || tr
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)"
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"sync-gbrain","questio
  • [bash] ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_id":"
  • [bash] ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","type":"
  • [bash] _TEL_END=$(date +%s)
  • [bash] ~/.claude/skills/gstack/bin/gstack-gbrain-detect 2>/dev/null
  • [bash] bun run ~/.claude/skills/gstack/bin/gstack-gbrain-sync.ts <user-args>
  • [bash] SOURCE_ID=$(grep -o '"source_id":"[^"]*"' ~/.gstack/.gbrain-sync-state.json 2>/d
  • [bash] SLUG="_capability_check_$$"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 19 inline command(s):
  • $ bash _UPD=$(~/.claude/skills/gstack/bin/gstack-update-check 2>/dev/null || .clau...
  • $ bash rm -f ~/.gstack/.writing-style-prompt-pending touch ~/.gstack/.writing-styl...
  • $ bash open https://garryslist.org/posts/boil-the-ocean touch ~/.gstack/.completen...
  • $ bash touch ~/.gstack/.telemetry-prompted
  • $ bash touch ~/.gstack/.proactive-prompted
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" 2>/dev/null |...
  • $ bash _GSTACK_HOME="${GSTACK_HOME:-$HOME/.gstack}" # Prefer the v1.27.0.0 artifac...
  • $ bash # Chosen mode: full | artifacts-only | off "$_BRAIN_CONFIG_BIN" set artifac...
  • $ bash "~/.claude/skills/gstack/bin/gstack-brain-sync" --discover-new 2>/dev/null ...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)" _PROJ="${GSTA...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-log '{"skill":"sync-gbrain","qu...
  • $ bash ~/.claude/skills/gstack/bin/gstack-question-preference --write '{"question_...
  • $ bash ~/.claude/skills/gstack/bin/gstack-learnings-log '{"skill":"SKILL_NAME","ty...
  • $ bash _TEL_END=$(date +%s) _TEL_DUR=$(( _TEL_END - _TEL_START )) rm -f ~/.gstack/...
  • $ bash ~/.claude/skills/gstack/bin/gstack-gbrain-detect 2>/dev/null
  • $ bash bun run ~/.claude/skills/gstack/bin/gstack-gbrain-sync.ts <user-args>
  • $ bash SOURCE_ID=$(grep -o '"source_id":"[^"]*"' ~/.gstack/.gbrain-sync-state.json...
  • $ bash SLUG="_capability_check_$$" if [ -f ~/.gbrain/config.json ] && \ gbrain ...
  • $ git pull
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Needs work skill with 11 suggestions for improvement (weakest: trust)

unfreeze

F
skill claude-native Score: 39%
completeness 63% weight 20%
clarity 57% weight 15%
actionability 50% weight 20%
safety 5% weight 15%
testability 45% weight 10%
trust 15% weight 20%
Verdict: block Profile: general Malice: medium Capability: high

block: profile=general, malice=medium, capability=high, categories=DESTRUCTIVE, EXFILTRATION, PRIVILEGE

Details & Suggestions

completeness (63%)

  • Has description
  • Has runbook body (126 words)
  • Has 2 code block(s) in body (inline examples)
  • Runbook style (no formal inputs)
  • Add more structure (sections or bullet lists) to the body
  • Add a dedicated ## Examples section for discoverability
  • Add gotchas/caveats to warn about common failure points

clarity (57%)

  • Has body content (126 words)
  • Language is concrete and specific
  • Description is too short — expand to 20-200 characters
  • Add section headers (##) and bullet lists to improve scannability

actionability (50%)

  • Runbook body present
  • Body references tools or includes code
  • No inputs defined (not always needed)
  • Add action-oriented instructions (bullet points starting with verbs)

safety (5%)

  • Add gotchas/caveats to warn about common failure points
  • Add specific gotchas (describe what can go wrong and why)
  • Address what happens when things go wrong (errors, failures, retries)

testability (45%)

  • Body contains inline examples
  • Body code blocks serve as examples
  • Add examples that describe expected outcomes

trust (15%)

  • Contains 2 executable code block(s): bash
  • [bash] mkdir -p ~/.gstack/analytics
  • [bash] eval "$(~/.claude/skills/gstack/bin/gstack-paths)"
  • Includes 31 companion script(s): analytics.ts, archetypes.ts, build-app.sh, build.sh, compare-pr-version.ts, detect-bump.ts, dev-skill.ts, discover-skills.ts, eval-compare.ts, eval-list.ts, eval-select.ts, eval-summary.ts, eval-watch.ts, garry-output-comparison.ts, gen-llms-txt.ts, gen-skill-docs.ts, host-config-export.ts, host-config.ts, jargon-list.json, models.ts, one-way-doors.ts, preflight-agent-sdk.ts, psychographic-signals.ts, question-registry.ts, setup-scc.sh, skill-check.ts, slop-diff.ts, task-emission-schema.ts, test-free-shards.ts, update-readme-throughput.ts, write-version-files.sh
  • Contains 2 inline command(s):
  • $ bash mkdir -p ~/.gstack/analytics echo '{"skill":"unfreeze","ts":"'$(date -u +%Y...
  • $ bash eval "$(~/.claude/skills/gstack/bin/gstack-paths)" STATE_DIR="$GSTACK_STATE...
  • CRITICAL: 5 suspicious pattern(s) found — review carefully before use
  • [DESTRUCTIVE] Hard reset destroys uncommitted work
  • [DESTRUCTIVE] DROP TABLE/DATABASE is destructive
  • [EXFILTRATION] May leak secrets (credential in output)
  • [EXFILTRATION] Logging credentials to console
  • [PRIVILEGE] Requests elevated privileges

Poor skill with 15 suggestions for improvement (weakest: safety)