
Acquire Codebase Knowledge
Maps an existing codebase into seven structured docs covering stack, structure, architecture, and risks.
- Repository
- github/awesome-copilot
- Stars
- 40,703
- Licence
- MIT
- Files
- 11 files
What it does
When you ask an AI agent to map or onboard you into a project, this skill makes it run a local scan script over the code, read any intent documents such as README or PRD files, and write seven markdown files into docs/codebase/ (stack, structure, architecture, conventions, integrations, testing, and concerns). Every claim is meant to cite a file path, and anything it cannot verify is marked TODO or as a question for you.
Great for
- Developers joining an unfamiliar repository who want a written orientation
- Tech leads who want a documented baseline of an existing project's architecture and risks
- Teams preparing a codebase for handover or audit
Skip it if
- Routine feature work or bug fixes, which the skill itself says it should not trigger for, because it adds documentation overhead
- Very large monorepos where the fixed scan limits (for example 200 tree entries and 3 directory levels) will miss much of the structure
- Projects where you need the docs to be authoritative without reviewing them, since the output is only as accurate as the files it reads
Try asking
- “Map this codebase and write the docs to docs/codebase”
- “Onboard me to this repo and focus on testing and concerns”
- “Document the architecture of this project”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 11 original files, the MIT licence and an attribution note.
acquire-codebase-knowledge.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip acquire-codebase-knowledge.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip acquire-codebase-knowledge.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Run it from the project root so the scan script and relative output path resolve correctly.
- Give a focus area such as 'architecture only' to get deeper coverage of one document first.
- Review the [ASK USER] questions in the final answer, since the skill cannot infer team intent from code alone.
What's inside
- assets/templates/ARCHITECTURE.md
- assets/templates/CONCERNS.md
- assets/templates/CONVENTIONS.md
- assets/templates/INTEGRATIONS.md
- assets/templates/STACK.md
- assets/templates/STRUCTURE.md
- assets/templates/TESTING.md
- references/inquiry-checkpoints.md
- references/stack-detection.md
- scripts/scan.py
- SKILL.md
Redistributed unmodified from github/awesome-copilot under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.