.claude/skills/core/project-scan/SKILL.md
Auto-populate context/project.md by reading the current codebase
npx skillsauth add andrem-sec/psc-comet project-scanInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Reads the current codebase and auto-populates context/project.md. Run once on setup, re-run when the project's structure changes significantly.
Without a scan, context/project.md stays as a template. Every skill that reads it gets nothing useful. The context layer is only as good as the data in it.
Check for manifest files in order:
| File | Indicates |
|------|-----------|
| package.json | Node.js — check dependencies for framework (next, express, fastapi via py) |
| pyproject.toml / setup.py / requirements.txt | Python — check for FastAPI, Django, Flask, etc. |
| go.mod | Go — check module name and key imports |
| Cargo.toml | Rust |
| pom.xml / build.gradle | Java/Kotlin |
| Gemfile | Ruby |
| composer.json | PHP |
| *.csproj | .NET/C# |
Read the manifest. Extract: language, framework, key dependencies (non-obvious ones only — not lodash, not requests).
| Stack | Look For |
|-------|---------|
| Node.js | "main" in package.json, src/index.ts, app.ts, server.ts |
| Python | main.py, app.py, __main__.py, manage.py (Django) |
| Go | cmd/*/main.go, main.go |
| Rust | src/main.rs, src/lib.rs |
Read the entry point file. Note what it initializes.
List directories at the top level of src/, lib/, app/, pkg/, or equivalent. For each:
index.ts, __init__.py, or equivalent, read the first 20 linesCap at 10 modules. If there are more, note the count and describe the most significant.
| File/Pattern | Indicates |
|-------------|-----------|
| docker-compose.yml | Local Docker services — read service names |
| Dockerfile | Containerized deployment |
| .github/workflows/ | GitHub Actions CI |
| *.tf files | Terraform infrastructure |
| k8s/ or kubernetes/ | Kubernetes deployment |
| .env.example | Read to understand required environment variables |
Read README.md (first 50 lines). Extract anything that describes architecture, constraints, or setup requirements not already captured.
Write directly to context/project.md, replacing template placeholders. Do not append — overwrite the template sections with real data.
Sections to populate:
State: "context/project.md populated. Please review and fill in:
Do not guess at framework from file extensions alone. Read the manifest.
Do not list every dependency — only the non-obvious ones that shape how the project works.
Do not overwrite anything the user has already filled in. If a section has real content (not template placeholder text), preserve it.
data-ai
Parallel agent swarm — decomposes work into independent units, spawns isolated workers, tracks PRs via fan-in
testing
Audit animations and transitions for motion accessibility, performance safety, and design intent. Enforces prefers-reduced-motion compliance and blocks layout-triggering transitions.
testing
Test specifically for AI-introduced regressions that repeat without tests
development
Framework for decomposing agent-driven tasks into independently verifiable units