sf-eval
Makes evals a precondition of work: per-surface suites signed off before stories queue, per-story fail_to_pass/pass_to_pass cases with stable IDs, deterministic-first grading, cost-tiered runs, sha256 receipts. Use for eval setup, story evaluation or failure triage.
- License
- MIT
- Compatibility
- Works with Claude Code, Cursor and Codex.
Pinned to revision f045d14f95c2, so it is the text this page describes rather than whatever the author pushed since.
Files
- skills/sf-eval/SKILL.md
- skills/sf-eval/examples/eval.config.example.json
- skills/sf-eval/references/cases.md
- skills/sf-eval/references/grading.md
- skills/sf-eval/references/modes.md
- skills/sf-eval/references/receipts.md
- skills/sf-eval/references/setup.md
- skills/sf-eval/scripts/sf_eval.py
Every link opens the file at its source, pinned to the revision this page describes.