
eval-harness
A framework for defining and running formal evaluations of Claude Code sessions to measure reliability and prevent regressions.
Lift+25ppscreen tier · directional
90
Trustsecurity · quality · license · upkeep
Scan Saferisk 3/100
LicenseMITserved via MCP
HEAD:2abb81705b8b Safe10/15/2026
Update tracking (upstream polling, re-scan, re-benchmark, accept-or-stay-pinned) ships with Phase 4.