Project Init Skill
Description
A Claude Code skill that runs a 3-round project kickoff interview before writing any code — outputs full architecture package including tech stack, DB schema, .env template, security checklist, and CLAUDE.md
Installation
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open the source below and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
README
project-init — Claude Code Project Kickoff Skill
A [Claude Code skill](https://docs.anthropic.com/en/docs/claude-code/skills) that forces the right architecture decisions **before writing a single line of code**.
Type `/project-init` at the start of any new project. Claude will ask 9 targeted questions across 3 rounds, then auto-generate a complete architecture package — tech stack, database schema, directory structure, `.env` template, security checklist, and a ready-to-use `CLAUDE.md`.
The Problem This Solves
Most AI-assisted projects go wrong because the AI starts coding immediately without asking:
- Is this local or public-facing?
- Will there be real users or just you?
- Does this need file uploads, OAuth, or third-party services?
The result: you build with SQLite and local file storage, then realize halfway through you need PostgreSQL and cloud storage. You refactor auth. You rewrite the deployment config. **This skill eliminates that entire class of rework.**
What It Does
3-Round Interview
| Round | Questions | Determines |
|---|---|---|
| 1 | Deployment target, user scale, data sensitivity | Database choice, security level |
| 2 | File uploads, notifications, third-party integrations | Storage solution, service config |
| 3 | Tech preferences, existing accounts, timeline | Stack selection, deployment platform |
9-Part Output Package
| # | Output | Description |
|---|---|---|
| 1 | PRD Summary | One-paragraph description — confirm Claude understood the project before proceeding |
| 2 | ADR (Architecture Decision Record) | Why each technology was chosen, what was rejected and why |
| 3 | Tech Stack | Every layer with specific versions and rationale |
| 4 | Database Schema | Main tables, fields, types, relationships |
| 5 | Directory Structure | Full project skeleton with annotated comments |
| 6 | .env Template |
All environment variables, grouped, wi |
Related Skills
Fastapi Review
Review a FastAPI application for architecture, async correctness, dependency injection, Pydantic schemas, secu
Defense in Depth
Implement multi-layered testing and security best practices.
SecLists Official Repository
[OWASP Testing Guide](https://owasp.org/www-project-web-security-testing-guide/)
Threat Hunting with Sigma Rules
Use Sigma detection rules to hunt for threats and analyze security events
Maintenance Walkthrough - 2026-03-29
- Re-triaged the full 2026-03-15 security finding set against current `main` and wrote a fresh current-head re
Google Workspace Model Armor
Filter user-generated content for safety
