A

Agent Friendly Code

Agent Skill GitHub Action Live Score Methodologyv0.7.0

← all packages

Package

npm/

prettier

Score

prettier/prettierGitHub

Full score breakdown

70.8

Signals are static heuristics — no agent is actually run. Per-model rationales are docs-cited; the weight values themselves are still pre-benchmark.

© 2026 Himanshu SinghMITPrivacyTerms

Home About Methodology Roadmap Changelog

Agent Skill GitHub Action Live Score Packages