Skip to content
kauankelvin7Public

About

Evidence-first GitHub repository health analyzer with deterministic scoring, GitHub App auth, signed webhooks and CI/CD.

Topics

Resources

Code of conduct

Contributing

Security policy

Stars

0 stars

Watchers

0 watching

Forks

Repository files navigation

RepoLens

Repository health, built on evidence.

Live CI CodeQL License

RepoLens analyzes a public GitHub repository and turns objective engineering signals into an evidence-first health report. Every score is tied to observable repository signals and real file references where available. It focuses on documentation, CI/CD, observable security controls, maintenance, and engineering quality without sending source code to an external AI model.

Private repositories are intentionally not exposed by the public analyzer; they will require an explicit user-authorization flow in a future version.

What it evaluates

  • README, license, contributing guide and project documentation
  • GitHub Actions workflows and test presence
  • Dependabot and CodeQL configuration
  • SECURITY.md, environment examples and lockfiles
  • Security-focused tests, security documentation and recognizable access-policy files
  • Issue and pull request templates
  • Repository activity and metadata quality
  • Language distribution and typed-language signals
  • Prioritized, explainable recommendations

The Security Controls pillar measures repository-level controls visible from public metadata and file paths. It is not a vulnerability scanner, penetration test, runtime assessment, or certification that an application is secure.

Evidence-first analysis

The UI is designed around one question:

Why did RepoLens reach this conclusion?

For each engineering category, RepoLens surfaces detected/missing signals and the repository files that support the result, such as README.md, test files, GitHub Actions workflows, SECURITY.md, Dependabot configuration and lockfiles.

This keeps the product deterministic and auditable: no hidden model decides whether a repository is "good".

Architecture

The analyzer normally uses only three GitHub REST API requests per repository:

  1. Repository metadata
  2. Language statistics
  3. Recursive Git tree

The analysis engine is deterministic and runs server-side.

Browser
  -> POST /api/analyze
  -> validation + rate limit
  -> GitHub REST API
  -> deterministic scoring engine
  -> evidence extraction
  -> evidence-first dashboard

See docs/ARCHITECTURE.md for trust boundaries and design decisions.

Local development

Requirements: Node.js 20.9+ and npm.

npm install
copy .env.example .env.local
npm run dev

A token is optional for public repositories, but GITHUB_TOKEN raises the GitHub API rate limit.

Quality gates

npm run lint
npm run typecheck
npm test
npm run build

GitHub Actions runs the same checks on pushes and pull requests. CodeQL and Dependabot configurations are included.

GitHub App

RepoLens is also deployed as a real GitHub App integration.

It includes:

  • RS256 GitHub App JWT generation
  • installation discovery and installation token primitives
  • HMAC SHA-256 webhook verification
  • an App configuration status endpoint
  • a signed webhook endpoint at /api/webhooks/github
  • repository-scoped cache invalidation on signed GitHub webhook events
  • production App authentication validation through /api/app/status

See docs/GITHUB_APP.md for the recommended permissions and setup.

Security

Credentials remain server-side. Never prefix secret variables with NEXT_PUBLIC_. See SECURITY.md.

License

MIT © 2026 Kauan Kelvin Santos Barbosa

About

Evidence-first GitHub repository health analyzer with deterministic scoring, GitHub App auth, signed webhooks and CI/CD.

Topics

Resources

Code of conduct

Contributing

Security policy

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages