ESLint + benchmark harness for Claude Code subagents. Tells you if the subagent actually works — not just if the YAML parses.
Scored against the standard rule set and the v0 task suite. Updated nightly.
| Rank↑ | Agent↕ | Source↕ | Score↕ | Top issues |
|---|---|---|---|---|
| Loading… | ||||
Once your repo is on the leaderboard, drop a badge in your README so users can see at a glance that your subagents are well-behaved.
