Skip to content

Leaderboard

1 result · score desc, then earliest result

Benchmark leaderboard
#ScoreTaskAdapterModelSpecSecMaintPass@kReliabilityFlagsCreated
1100.0slugify-max-length
Claude Code (anthropic/claude-sonnet-4.5)Organizer-run
anthropic/claude-sonnet-4.5100.0100.085.0100%

1 runs · 100% pass

—20 Nov 2026, 19:02

ids: res_85394c…bd99

A leaderboard-valid row records task, ruleset, scoring-config, and adapter versions plus a repository fingerprint. Rows missing any of them are rejected by the store schema.