Metadata-Version: 2.4
Name: braingent
Version: 0.1.1
Summary: Durable engineering memory CLI: doctor, reindex, find, recall, synthesize, MCP, and QA test-plan generation.
Project-URL: Homepage, https://github.com/thedoublejay/braingent-manifesto
Project-URL: Source, https://github.com/thedoublejay/braingent-manifesto
Project-URL: Issues, https://github.com/thedoublejay/braingent-manifesto/issues
Author: JJ Adonis
License-Expression: MIT
License-File: LICENSE
Keywords: cli,mcp,memory,qa,test-plans
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Software Development
Requires-Python: >=3.11
Requires-Dist: pyyaml<7,>=6.0.3
Provides-Extra: dev
Requires-Dist: build>=1.5.0; extra == 'dev'
Requires-Dist: hatchling>=1.30.1; extra == 'dev'
Requires-Dist: mypy>=2.1.0; extra == 'dev'
Requires-Dist: pytest>=9.1.0; extra == 'dev'
Requires-Dist: ruff>=0.15.17; extra == 'dev'
Requires-Dist: twine>=6.2.0; extra == 'dev'
Requires-Dist: types-pyyaml>=6.0.12.20260518; extra == 'dev'
Provides-Extra: mcp
Requires-Dist: mcp<2,>=1.27.2; extra == 'mcp'
Provides-Extra: tokens
Requires-Dist: tiktoken<1,>=0.13.0; extra == 'tokens'
Description-Content-Type: text/markdown

# Braingent Memory Repo

This is a Markdown-first engineering memory repository based on Braingent, created by JJ Adonis.

Use it as a durable memory layer for AI-assisted software engineering work. Claude, Codex, ChatGPT, and other tools should read this repo before planning meaningful work and should capture important outcomes after the work is done.

## What Goes Here

- Project and repository profiles.
- Task records.
- Code review records.
- Decision records.
- Reusable learnings.
- Tool and version notes.
- Ticket stubs for cross-cutting work.
- Optional live `BGT-NNNN` agent tasks for active coordination.
- Optional dashboard docs for a read-only live task UI.
- An installable `braingent` helper CLI for search, validation, reindexing,
  task files, MCP retrieval, and QA generation.
- The packaged QA generator for Markdown, Xray JSON, TestRail CSV, and Gherkin
  outputs.
- Raw imports before they are summarized.
- Preferences that guide future AI agents.

## Read First

Agents should start here:

1. `AGENTS.md` or `CLAUDE.md`, depending on the tool.
2. `INDEX.md`.
3. `CURRENT_STATE.md`.
4. `preferences/`.
5. `tasks/INDEX.md` if live tasks are enabled and the work may already be active.
6. Relevant organization, project, repository, topic, tool, ticket, or person records.

## Core Workflow

Before work:

- Read the root instructions.
- Search memory for relevant context.
- Check live tasks before creating overlapping active work.
- Reuse prior decisions and conventions.

During work:

- Track decisions, versions, commands, failures, fixes, tickets, PRs, branches, and follow-ups.
- Append activity to a live `BGT-NNNN` task when coordination or handoff matters.

After work:

- Create or update a durable record.
- Link completed live tasks to durable records with `agent_task: BGT-NNNN`.
- Update indexes or current state if needed.
- Commit the memory change.

## Search

Start with free-text search:

```bash
rg -n "<query>" .
```

Use the packaged structured search helper:

```bash
braingent find kind=decision
braingent recall repo=repo--example--owner--repo
```

You can also search frontmatter fields directly:

- `record_kind`
- `status`
- `organization`
- `project`
- `repositories`
- `ticket`
- `topics`
- `tools`
- `people`
- `ai_tools`

## Safety

Never store secrets, credentials, tokens, private keys, customer secrets, or sensitive personal data in this repo.

Use placeholders and links instead of copying sensitive evidence.
