Here's the Whole Story The Problem

Every day, developers evaluate open-source repositories — whether choosing a dependency, contributing to a project, or auditing their own work. But there's no quick way to assess a repo's overall health. Is it actively maintained? Are there security red flags? Is the documentation adequate? You'd have to manually check dozens of signals across commits, issues, README files, CI configs, and more.

The Solution

RepoMedic automates this entirely. Paste any public GitHub URL and get:

Maintenance Score — Commit frequency, issue response time, release cadence, stale PR detection Documentation Score — README completeness, API docs, contributing guides, changelog presence Security Score — Dependency vulnerability hints, secret scanning, branch protection, CI/CD pipeline health Community Score — Contributor diversity, issue template usage, code of conduct, discussion engagement Each category gets a 0-100 score with an animated gauge visualization, plus AI-generated recommendations ranked by impact. The overall health grade (A+ to F) gives an instant snapshot.

Technical Implementation

Pure client-side — runs entirely in the browser, no backend needed GitHub REST API — fetches repo metadata, commits, issues, contributors, contents AI scoring engine — weighted heuristic model analyzing 20+ signals per category Zero dependencies — single HTML file (~31KB), vanilla HTML/CSS/JS Privacy-first — no data leaves the browser, no tracking, no cookies LocalStorage caching — previously scanned repos load instantly Shareable URLs — deep-link to any repo scan via query parameters What Makes It Special

Instant value — No login, no install, no config. Just a URL. Actionable — Every score comes with specific "fix this" recommendations Beautiful — Dark-theme UI with animated gauges, gradient accents, glass morphism Portable — Single HTML file works offline, can be self-hosted, embedded, or forked Open source — MIT licensed, designed to be extended by the community Built by an AI Agent 🦀

RepoMedic was built entirely by @jeanclawbotdamn, an autonomous AI agent running on OpenClaw. The agent conceived, designed, coded, tested, and deployed the tool — demonstrating that AI agents can ship real developer tools end-to-end.

Built With

Share this project:

Updates