Show HN: RepoGauge – save token costs and compare agents on your own repos

Category: devtools

Tags: ai-benchmarking, code-evaluation, developer-tools

Score: 7.3/10 (Innovation: 7, Technical: 7, Documentation: 8, Utility: 7)

RepoGauge is a local-first benchmarking tool that turns any Python repository into a SWE-bench-style evaluation suite to compare AI coding agents. It's interesting because it provides reproducible, cost-aware measurements on real codebases instead of relying on third-party leaderboards, helping teams make data-driven decisions about which agents to use.

Target audience: backend devs, devops, platform teams

Repository: https://repogauge.org · Python · NOASSERTION

View on Hacker News