fd1d86ba0b
- SKILL.md frontmatter: add Name, Tier, Category, Dependencies, Author, Version as capitalized top-level keys (validator requirement) - SKILL.md sections: add Name and Description headings (validator requirement) - Add expected_outputs/ with 3 sample files: PRD README, page doc, enum dict - prd_scaffolder.py: add validate_analysis(), --validate-only, --dry-run flags, structured print_summary() — now 333 LOC (was 255, within 300-500) - Add scripts/.gitignore to exclude generated prd/ test output Scores: validator 65→97.6 (EXCELLENT), quality 51→73.2 (B-), scripts 2/2 PASS Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Code → PRD
Reverse-engineer any codebase into a complete Product Requirements Document (PRD).
Quick Start
# One command
/code-to-prd /path/to/project
# Or step by step
python3 scripts/codebase_analyzer.py /path/to/project -o analysis.json
python3 scripts/prd_scaffolder.py analysis.json -o prd/ -n "My App"
Supported Frameworks
| Stack | Frameworks |
|---|---|
| Frontend | React, Vue, Angular, Svelte, Next.js, Nuxt, SvelteKit, Remix |
| Backend | NestJS, Express, Django, DRF, FastAPI, Flask |
| Fullstack | Next.js (pages + API), Nuxt (pages + server), Django (views + templates) |
What It Generates
prd/
├── README.md # System overview
├── pages/
│ ├── 01-user-mgmt-list.md # Per-page/endpoint docs
│ └── ...
└── appendix/
├── enum-dictionary.md # All enums and status codes
├── api-inventory.md # Complete API reference
└── page-relationships.md # Navigation and data coupling
Scripts
| Script | Purpose |
|---|---|
codebase_analyzer.py |
Scan codebase → extract routes, APIs, models, enums |
prd_scaffolder.py |
Generate PRD directory skeleton from analysis JSON |
Both are stdlib-only — no pip install needed. Run --help for full usage.
References
references/framework-patterns.md— Route, state, API, form, and model patterns per frameworkreferences/prd-quality-checklist.md— Validation checklist for completeness and accuracy
Attribution
Inspired by code-to-prd by @lihanglogan.
License
MIT