loading
loading
Generate a behavior eval suite for a skill — trigger, anti-trigger, output-contract, and failure cases.
Complements the eval-harness runner: this one writes the cases, including the anti-triggers and failure paths most authors never test.
Listed for review
No verified public repo for this skill yet, so this page does not give you an install command. Skills with a verified source install in one command — or fully manual: copy the skill folder into .claude/skills/ and your agent picks it up.
Boostor Quality Score
84/100 · B
Skill Eval Writer reads a SKILL.md and its known failure modes, then writes an eval suite that actually tests behavior: positive cases that should trigger the skill, negative cases that should not, output-contract checks, edge cases, and regression prompts. It aims for at least eight evals covering trigger, anti-trigger, the output shape, and failure handling — and it avoids writing tests that hand the model the answer or only ever exercise the happy path.
Build repeatable LLM evals with golden datasets and scoring rubrics.
Most teams ship prompt changes on vibes. This makes model quality a CI gate with golden sets and rubrics, so a regression fails the build like any bug.
Transparent + deterministic: every point above is computed from this skill's real fields plus a prompt-injection safety scan. No black box, no pay-to-rank.