| name | onboard-codebase |
| description | Analyze existing web development codebase and generate documentation for AI-assisted development. Use when adding harness to existing projects (1→N). |
| user-invocable | true |
| argument-hint | |
| allowed-tools | Read, Glob, Grep, Write, Bash, WebSearch |
/onboard-codebase - Analyze Existing Project
Analyze existing codebase and generate AI-ready documentation.
Usage
/onboard-codebase
When to Use
- Adding harness to existing project
- After major refactoring
- When onboarding new team members
- Before starting significant changes
Analysis Process
1. Project Detection
Detect:
- Frontend: React/Vue, component structure, state management
- Backend: FastAPI/Flask, API structure, middleware
- Database: PostgreSQL schemas, migrations, Redis usage
- Deployment: Docker, CI/CD pipelines
2. Architecture Mapping
Generate:
- Component hierarchy diagram
- API endpoint catalog
- Database relationship map
- Service boundaries (for microservices)
3. Documentation Generation
Create:
docs/architecture/overview.md - System architecture
docs/api/endpoints.md - API documentation
docs/database/schema.md - Database schemas
CLAUDE.md - Project context for AI
4. Agent Context
Populate agent knowledge:
- Frontend patterns used
- Backend conventions
- Testing setup
- Deployment procedures
Output
✅ Codebase onboarded
Generated:
- docs/architecture/overview.md
- docs/api/endpoints.md
- docs/database/schema.md
- CLAUDE.md (updated)
Detected:
- Frontend: React with TypeScript
- Backend: FastAPI
- Database: PostgreSQL + Redis
- Deployment: Docker + GitHub Actions
Next: Run /start to begin AI-assisted development
Reference
Base implementation: @references/base/progress/skills/onboard-codebase/SKILL.md