project-intelligence

name	project-intelligence
description	Map an unfamiliar codebase, identify architecture, conventions, commands, entrypoints, dependencies, risks, and a safe implementation strategy. Use before non-trivial changes or when joining a new repository.

Project Intelligence

Use this skill to build a compact mental model of a repository before changing it.

Workflow

1. Identify repository shape:

   • pwd, ls, find . -maxdepth 3 -type f or rg --files
   • package/build files: package.json, Cargo.toml, go.mod, pyproject.toml, Makefile, CI configs
   • context files: AGENTS.md, CLAUDE.md, .opencode*, .pi, .github

2. Determine commands:

   • test, lint, typecheck, build, format
   • prefer package scripts over guessing raw commands
   • note commands you did not run and why

3. Map architecture:

   • entrypoints and top-level modules
   • core domain types/data flow
   • config/environment assumptions
   • test layout and fixtures

4. Find local conventions:

   • naming/style patterns
   • error handling and logging
   • dependency injection / state management
   • existing examples similar to the requested change

5. Produce a concise report:

## Project Map
- Stack:
- Entrypoints:
- Important directories:
- Commands:
- Conventions:
- Risks/unknowns:
- Suggested implementation approach:

Rules

• Prefer read over shelling out to cat for file contents.
• Read important files completely when feasible.
• Do not modify files while using this skill unless the user explicitly asks to implement.
• Ask one clarifying question if the repo goal or requested change is ambiguous.