AgentRuleBench
A reproducible benchmark to test AI coding agents for architectural
1 ratings4.14
Saves2
CategoryDeveloper Tools
PlatformWeb
PricingFree
AgentRuleBench is a benchmark project that tests whether AI coding agents violate architectural import-boundary rules inferred from real codebases. It includes the testing harness, pre-registration documents, raw run data, and scoring tools to reproduce and verify results regarding compliance with specific import boundaries in TypeScript projects.
aiopen-source
Ratings & Reviews
Write a reviewBe the first to share a thoughtful review.
Information
CategoryDeveloper Tools
PlatformsWeb
PricingFree
AccountNot confirmed
ListedSeptember 2026
WebVisit
Report a problem
Around the Web
Show HN2 source points · 0 comments