The first leaderboard that measures what you BUILD

Ranked by output
not by spend.

Other leaderboards rank you by tokens consumed. ccrank measures your real developer output — commits, lines shipped, code quality, consistency — and ranks you against every Claude Code developer on the planet.

$npx ccrank

Velocity

Commits shipped per day

Efficiency

Output per token used

Impact

Scope & complexity

Consistency

Active days & streaks

Global Leaderboard

Rank
Developer
Overall
Velocity
Efficiency
Impact
Consistency
Tier

How it works

One command. Real scores. No bullshit.

Step 1

Run the CLI

One command. That's it. The CLI scans your Claude Code sessions and git history locally.

$ npx ccrank
Step 2

We analyze your output

Commits, lines shipped, files created, languages used, active days, streaks — all from your local git repos.

Step 3

Score calculated

Your score is based on 4 dimensions: Velocity, Efficiency, Impact, and Consistency. Not token spend.

Step 4

Get your rank

See your global percentile, tier badge, and how you stack up against other Claude Code developers.

The Score Formula

// Your overall score (0-1000) overallScore = velocity × 0.25 // commits per active day + efficiency × 0.30 // lines shipped per token + impact × 0.25 // weighted file & line changes + consistency × 0.20 // active days + streak bonus

Efficiency is weighted highest (30%) — we reward developers who get MORE done with LESS tokens.