Statically checks harness assets against a known-answer probe set after rule/skill changes — inspects the changed source for the trigger phrases, chain links and gate conditions each probe expects, and reports PASS/FAIL/SKIP per probe. Source inspection only; it does not run live sessions, so it catches assets that no longer SAY the right thing, not models that stop DOING it.
Install
npx skills add chrono-meta/forge-harness --skill prompt-regressionnpx skills add chrono-meta/forge-harness --skill prompt-regression -a claude-code -gnpx skills add chrono-meta/forge-harness --skill prompt-regression -a codex -ggemini skills install https://github.com/chrono-meta/forge-harness.git --path plugins/fh-meta/skills/prompt-regressiongh skill install chrono-meta/forge-harness prompt-regressiongit clone --depth 1 https://github.com/chrono-meta/forge-harness /tmp/chrono-meta__forge-harness
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/chrono-meta__forge-harness/plugins/fh-meta/skills/prompt-regression ~/.claude/skills/prompt-regression # Claude Code
cp -R /tmp/chrono-meta__forge-harness/plugins/fh-meta/skills/prompt-regression ~/.agents/skills/prompt-regression # Codex, Cursor, Gemini CLI, Copilot, OpenCode