Ranking Eval Loop
Description
# Ranking eval loop — mine real misses, then measure the fix A ranking change is the easiest kind of change to ship on a hunch and the hardest to be right about, because a re-ranked top-10 looks plausible either way. This loop makes that impossible: the misses come from real sessions, the labels are written before the ranker is run, and nothing ships without a held-out delta. Read `docs/EVALS.md` §1 (the instruments) and §4 (ranking changes, measured) before you start. ## Step 0 — recalibrate
Installation
Installs to ~/.claude/skills/redhat-et-ripwire-ranking-eval-loop/SKILL.md
mkdir -p ~/.claude/skills/redhat-et-ripwire-ranking-eval-loop && curl -fsSL https://raw.githubusercontent.com/redhat-et/ripwire/HEAD/prompts/ranking-eval-loop.md -o ~/.claude/skills/redhat-et-ripwire-ranking-eval-loop/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
Full documentation available on GitHub
View Source RepositoryRelated Skills
Auto Update
Pull the latest ECC repo changes and reinstall the current managed targets.
Ecc Guide
Navigate ECC's current agents, skills, commands, hooks, install profiles, and docs from the live repository su
Epic Claim
Claim an epic issue, stamp coordination state, and sync local ownership.
Epic Publish
Publish a validated epic update back to the issue and local cache.
Epic Review
Mark epic review requested, approved, or changes requested.
Epic Unblock
Sweep blocked epic issues and reopen anything whose dependencies are closed.