← ClaudeAtlas

forge-evallisted

Run Stage 7 of the Forge pipeline — evaluation and quality assurance. Use when the user says /forge:eval, wants to verify the implementation, check test coverage, or validate against requirements. Requires Stage 1–6. Invokes the evaluator.
tonmoy007/forge-plugins · ★ 1 · AI & Automation · score 75
Install: claude install-skill tonmoy007/forge-plugins
# /forge:eval — Evaluation ## When to Use - User says `/forge:eval` - User wants to verify the implementation, check test coverage, or validate against the SRS - Working in a Forge project at Stage 6 or 7 ## Pre-flight Check **Entry gate (REQ-GATE-ENTRY-001)** — before adopting the persona, verify the prior stage's artifact exists: ```bash python3 ${CLAUDE_PLUGIN_ROOT}/scripts/state-manager.py preflight --stage 7 ``` If it exits non-zero, **STOP**: present its message verbatim and do not proceed — the prior stage must be completed first (or use `/forge:force-advance` to skip intentionally). 1. Read `pipeline/state.md` — confirm Forge project. 2. Confirm `pipeline/01-srs/srs.md` and `pipeline/04-spec/technical-spec.md` exist. 3. Run `python3 ${CLAUDE_PLUGIN_ROOT}/scripts/check-gate.py --stage 6` to see Stage 6 status. If critical criteria fail, warn the user before proceeding. 4. Run `python3 ${CLAUDE_PLUGIN_ROOT}/scripts/load-profile.py --cwd . --stage 7` to load project-type evaluation criteria. **Every** profile criterion in `additional_criteria` (e.g., ML: model accuracy, GPU memory, drift detection; API: contract tests, load test, auth bypass; fullstack: Lighthouse, WCAG AA, responsive breakpoints; CLI: --help text, exit codes; library: API surface, semver, doc coverage) **must appear as a row in the eval matrix** with a pass/fail verdict and evidence. ## Steps 1. Read `agents/evaluator.md` to load the Evaluator persona. 2. Adopt that persona — you are now the