GhostlyGawd

Run Evals — Development skill for Claude Code

Development community

Replay the regression corpus inside THIS interactive session — no headless, no API key (ADR 0003).

How to install Run Evals

Installs to ~/.claude/skills/ghostlygawd-recursive-harness-run-evals/SKILL.md

Terminal

mkdir -p ~/.claude/skills/ghostlygawd-recursive-harness-run-evals && curl -fsSL https://raw.githubusercontent.com/GhostlyGawd/recursive-harness/HEAD/commands/run-evals.md -o ~/.claude/skills/ghostlygawd-recursive-harness-run-evals/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.

What Run Evals does

description: Replay the regression corpus inside THIS interactive session — no headless, no API key (ADR 0003). Required before merging enforcement-layer changes; part of /meta-retro.

provenance: 2026-06-12, user correction "there should be no headless" (ADR 0003); 2026-06-13 retro (session 56295237 → 61f58113) added the MSYS /tmp caution after a verified bash↔Windows-python path split; 2026-06-19 (followups 1639c1/dff31d, session 85bf58c5) extended the cygpath -w sandbox path to step 3

Alternatives in Development

Continuous Claude — Run Claude Code in a continuous loop, autonomously creating PRs, waiting for checks, and merging 1.3k ★
Agentic Engineering Handbook — The definitive OpenAI, Claude, MCP, Harness, Evals, and Production Agent Systems learning roadmap 553 ★
ImportMemory — You are merging an imported profile into a structured personal memory system 250 ★