repo-onboarding

name	repo-onboarding
description	Quickly map an unfamiliar repository — entrypoints, stack, package manager, conventions, architecture. Trigger on /repo-onboarding, or when entering a new repository, reviewing an unfamiliar codebase, or needing a project summary.
user-invocable	true
allowed-tools	Read, Grep, Glob, Bash

Repo Onboarding

Map an unfamiliar repository fast. Reduces guessing and repetitive questions when entering a new codebase.

Workflow

1. Entry points

• package.json, pyproject.toml, Cargo.toml, go.mod, etc. — identify package manager and dependencies.
• tsconfig.json, tsup.config.ts, vite.config.ts, next.config.js, etc. — identify build system.
• Dockerfile, docker-compose.yml, .github/workflows/ — identify deployment/CI.
• Directory structure at root level.

2. Stack detection

• Language (TypeScript, Python, Rust, Go, etc.)
• Framework (Next.js, Express, FastAPI, etc.)
• Database (PostgreSQL, SQLite, MySQL, etc.)
• Testing framework (Vitest, Jest, pytest, etc.)
• Linter/Formatter (oxlint, ESLint, Prettier, etc.)
• Monorepo tool (pnpm workspaces, turborepo, nx, etc.)

3. Conventions

• Naming conventions (files, functions, components, routes).
• Module/export pattern (ESM, CJS, barrel exports).
• Error handling pattern (try/catch, Result type, error middleware).
• Testing conventions (where tests live, naming, mocking approach).

4. Architecture summary

• High-level architecture (monolith, microservices, layered, feature-based).
• Key modules and their responsibilities.
• Data flow (request → handler → service → database → response).
• External dependencies and integrations.

5. Common commands

• dev, build, test, lint, format, typecheck.
• Any project-specific scripts (migration, seed, deploy).

Output

## Repo: <name>

**Stack**: ...
**Entry**: ...
**Commands**: ...

## Architecture
...

## Conventions
...

## Key Directories
...

Rules

• Read config files first — they contain the most reliable information.
• Do not run install or build commands during onboarding unless necessary.
• Output is for the current agent session — does not persist unless saved.

Experience Log

This skill learns from experience. Mechanism:

• Read .claude/experience/'"$s"'.md at skill start to benefit from past lessons.
• Write to .claude/experience/'"$s"'.md after use if you learned something new.
• Format: YYYY-MM-DD: <lesson> — one line per entry.
• Evolve: If the same issue repeats 3+ times, update this SKILL.md itself.