Ranked by what you BUILD, not what you spend.
Other leaderboards rank you by token consumption. ccrank measures your real developer output — commits shipped, lines of code, efficiency per token, consistency — and ranks you against every Claude Code developer on the planet.
npx ccrankThat's it. The CLI:
- Scans your git repos for commits, lines, files created
- Reads Claude Code usage data from
~/.claude/ - Calculates your score across 4 dimensions
- Submits to the global leaderboard
Your score (0-1000) is based on:
| Dimension | Weight | What it measures |
|---|---|---|
| Velocity | 25% | Commits shipped per active day |
| Efficiency | 30% | Lines of code per token consumed |
| Impact | 25% | Scope: files created, repos, languages |
| Consistency | 20% | Active coding days + streak |
Efficiency is weighted highest — we reward developers who get MORE done with LESS tokens.
| Tier | Percentile |
|---|---|
| Legendary | Top 1% |
| Diamond | Top 5% |
| Platinum | Top 15% |
| Gold | Top 30% |
| Silver | Top 60% |
| Bronze | Everyone else |
npx ccrank # Analyze last 30 days (default)
npx ccrank --period 7 # Last 7 days
npx ccrank --period all # All time
npx ccrank --dry-run # Calculate score without submitting
npx ccrank --repos ~/projects # Scan specific directory
npx ccrank --verbose # Show debug info- Clone the repo
- Copy
.env.exampleto.envand fill in:- PostgreSQL connection string (Neon free tier works great)
- GitHub OAuth app credentials
- NextAuth secret
- Run:
npm install npx prisma db push npm run dev
| ccrank | Viberank | CCgather | |
|---|---|---|---|
| Measures | What you built | Tokens spent | Tokens spent |
| Score based on | Commits, lines, efficiency | Token count | Token count |
| Rewards | Efficiency (less tokens = better) | More spending | More spending |
| Git analysis | Yes | No | No |
- Frontend: Next.js 15, Tailwind CSS, Framer Motion
- Database: PostgreSQL via Prisma
- Auth: NextAuth v5 with GitHub OAuth
- CLI: Pure Node.js, zero dependencies
MIT