Analyze any repository
like a recruiter would.
Get a deterministic 0–100 score across documentation, commits, tech stack, and architecture — with actionable improvement suggestions.
What gets analyzed?
Four independent analyzers evaluate your repository across the dimensions that matter most to technical recruiters and senior engineers.
README Quality
25%Checks for installation instructions, usage examples, badges, screenshots, architecture docs, and code blocks.
Commit Discipline
20%Analyzes temporal spread, conventional commit usage, message quality, and detects burst/duplicate patterns.
Tech Stack Maturity
30%Detects frameworks, testing, linting, TypeScript, CI/CD, containerization, and state management setup.
Architecture Complexity
25%Evaluates folder depth, separation of concerns, modularity (Shannon entropy), and penalizes anti-patterns.
How it works
From URL to actionable report in under 30 seconds.
Paste a GitHub URL
Enter any public GitHub repository URL and click analyze.
Parallel data fetching
6 parallel GitHub API calls fetch metadata, commits, tree, README, languages, and package.json.
Deterministic scoring
4 pure-function analyzers produce a weighted 0–100 score. Same repo = same score, always.
Actionable report
Get findings, strengths, suggestions, risk flags, and optionally an AI-enhanced improvement roadmap.
Hiring confidence levels
Your score maps to a confidence level that reflects how a technical recruiter would perceive the project.
Project needs significant work before it demonstrates engineering quality. Missing key signals like testing, proper documentation, or structured architecture.
Decent engineering fundamentals with clear room for improvement. Shows some good practices but lacks consistency across all dimensions.
Production-quality engineering signals across most categories. Well-documented, well-tested, well-structured, and professionally committed.
100% Deterministic
Same repository data always produces the same score. No randomness, no AI in the scoring pipeline.
Fully Explainable
Every point earned or lost traces back to a specific metric with a specific threshold. No black boxes.
AI-Enhanced (Optional)
Google Gemini optionally rewrites suggestions into a prioritized improvement roadmap. AI never affects scoring.