| name | rstack-skill-evaluator |
| description | Benchmark agent skills by generating eval cases, comparing skill-guided vs baseline runs, and recording artifacts under skills-test/{skill-name}. |
| metadata | {"dependencies":["skill-creator"],"internal":true} |
rstack-skill-evaluator
A thin repo-specific layer on top of skill-creator. For workflow (Test / Improve / Benchmark modes), JSON schemas, grading, and viewer details, defer to skill-creator's own SKILL.md and references/schemas.md.
Targeting a skill
If the user hasn't named a target, ask. Skills live under skills/ (production) and .agents/skills/ (internal-only).
Artifact layout
For skill <name>, two paths are tracked in git; everything else under skills-test/ is gitignored:
+--------------------------------------+----------------------------------+
| Tracked path | Purpose |
+--------------------------------------+----------------------------------+
| skills-test/<name>/evals/evals.json | eval definitions |
| skills-test/<name>/report.md | human-readable run summary |
+--------------------------------------+----------------------------------+
Workspaces, raw run outputs, and fixtures may live anywhere — under skills-test/<name>/ or an OS scratch dir — as long as report.md references the path so a reader can find them.
report.md
Committed Markdown summary of the latest run. At minimum cover: setup (model, skill version / commit ref, date), aggregate pass rate / tokens / wall time for with_skill vs without_skill, per-eval breakdown, and pointers to the raw artifacts.
