Appropriate use
Use for Slack updates, stand-ups, end-of-day reports, manager summaries, or team handoffs.
Produces a concise daily or stand-up update from verified work records without inventing progress, hiding blockers, or exposing restricted detail.
This page defines the behaviour, controls, evidence, and outputs required from the skill. It does not claim that a language model can enforce security or correctness by wording alone.
Use for Slack updates, stand-ups, end-of-day reports, manager summaries, or team handoffs.
Do not use as a substitute for source reports, performance evaluation, or an automatic broadcast without review.
Each phase has explicit inputs and outputs. Approval gates are hard stops bound to exact versions and hashes.
Confirm date, timezone, team, products, source systems, included work, private channels, and publication audience.
Collect ticket updates, test runs, defects, fix verification, automation changes, rollout checks, incidents, and environment issues from authoritative records.
Verify that claimed tests have results, build identity, evidence links, and current status. Separate planned, started, completed, and blocked work.
Select updates that affect release confidence, customer impact, team dependencies, decisions, or next-day work. Avoid activity-count noise.
Create concise sections for completed work, findings, blockers, release status, automation, next actions, and help needed.
Remove secrets, restricted customer detail, unsupported conclusions, blame, and internal evidence links inappropriate for the audience.
Present the source-backed draft and unresolved ambiguities. Bind approval to the selected time window and source snapshot.
Post or save idempotently, record the published text and source references, and avoid editing historical summaries without a visible correction.
An approval cannot be reused after its environment, source, action plan, or approved document changes.
| Gate | Decision | Required binding |
|---|---|---|
| 1 | Gate 1 · approve sources | Approval stores approver, role, timestamp, context hash, approved document hash, expiry, and invalidation state. |
| 2 | Gate 2 · approve summary | Approval stores approver, role, timestamp, context hash, approved document hash, expiry, and invalidation state. |
| 3 | Gate 3 · publish | Approval stores approver, role, timestamp, context hash, approved document hash, expiry, and invalidation state. |
Attempt outcomes remain separate from these workflow results. Errors, cancellations, and invalid context are never silently converted into a pass.
The summary is current, source-backed, audience-safe, and complete for the agreed scope.
Known source or coverage gaps are explicitly stated.
Material ambiguity, privacy concern, or unsupported claim remains.
No meaningful change occurred in the selected window.
Authoritative records disagree and require resolution.
Evidence has provenance, capture context, privacy classification, retention, and integrity metadata. A convenient artifact is not automatically sufficient proof.
| Claim | Primary evidence | Required validation |
|---|---|---|
| Completed work | Ticket or execution record | Status and build are current. |
| Defect | Issue link and triage status | Severity and release impact are not guessed. |
| Blocker | Owner and next action | The blocker is current and actionable. |
| Release | Release or rollout decision | Candidate and stage are identified. |
| Automation | PR, CI, or audit record | Progress describes verified change, not effort alone. |
| Next action | Owned plan | Action, owner, and dependency are concrete. |
These controls belong in the runtime, connector permissions, sandbox, renderer, storage layer, and review process.
Missing source data remains missing.
Stale ticket labels do not override newer verified reports.
Customer, security, incident, and credential details follow channel policy.
The summary describes facts, ownership, and next actions.
The skill drafts by default and posts only after explicit approval.
Material corrections are visible rather than silently replacing history.
The first applicable high-severity outcome takes precedence. A manual override is recorded separately and never rewrites the calculated result.
Resolve inconsistent authoritative records before publication.
Privacy, accuracy, or scope needs user confirmation.
Publish with explicit source or coverage limitations.
Publish a minimal no-change note only when useful.
Publish the concise verified summary.
The workflow documents what it cannot prove and what further evidence would change confidence.
Human summaries are generated from validated structured records. They are not maintained as separate, drifting sources of truth.
summary-context.jsonVersioned output produced by the workflow.
source-index.yamlVersioned output produced by the workflow.
activity-ledger.jsonlVersioned output produced by the workflow.
draft.mdVersioned output produced by the workflow.
redaction-review.mdVersioned output produced by the workflow.
approval.jsonVersioned output produced by the workflow.
published-message.mdVersioned output produced by the workflow.
daily-summary.htmlVersioned output produced by the workflow.
artifacts/qa/qa-daily-summary/<run-id>/ ├── summary-context.json ├── source-index.yaml ├── activity-ledger.jsonl ├── draft.md ├── redaction-review.md ├── approval.json ├── published-message.md └── daily-summary.html
Prepare my QA daily update for Bangkok time. Include SCOS 3.1.14 testing, PIXI rollout, the webOS issue, automation progress, blockers, and tomorrow's plan.
QA Daily Update Completed - Finished the required SCOS 3.1.14 OTA and reboot checks on the current candidate. - Verified the PIXI staged-rollout health snapshot at 20%. Findings - The webOS 6 repeat-launch black-screen issue remains reproducible on one model and is under triage. Automation - Studio Player E2E and benchmark work reached the verified milestone recorded in the project tracker. Blocker - One required SCOS image asset is still unavailable. Next - Resume the blocked SCOS path when the asset is published. - Continue model comparison for the webOS issue.
The written skill is only one layer. The repository, runtime, connectors, evidence storage, and publication path must implement these requirements.
Produces a concise daily or stand-up update from verified work records without inventing progress, hiding blockers, or exposing restricted detail.