Agent-Skills.md

Agent Skills: review-evals

Generate and open interactive HTML viewer for skill evaluation results. Use when reviewing eval output or inspecting grading details after a run.

UncategorizedID: bluewaves-creations/bluewaves-skills/review-evals

Repository

bluewaves-creations/bluewaves-skills
bluewaves-creationsLicense: MIT
1

Install this agent skill to your local

pnpm dlx add-skill https://github.com/bluewaves-creations/bluewaves-skills/tree/HEAD/plugins/skills-factory/skills/review-evals

Skill Files

Browse the full folder contents for review-evals.

Download Skill

Loading file tree…

plugins/skills-factory/skills/review-evals/SKILL.md

Skill Metadata

Name
review-evals
Description
Generate and open interactive HTML viewer for skill evaluation results. Use when reviewing eval output or inspecting grading details after a run.

Open the interactive HTML eval viewer for a skill's evaluation results.

$ARGUMENTS

If $ARGUMENTS is provided, use it as the skill path. Otherwise, ask the user which skill to review.

Steps

  1. Locate the skill directory and its .skill-eval/ workspace. Resolve $ARGUMENTS to an absolute path if provided.

  2. Find the latest run directory (or ask which run to review).

  3. Run generate_review.py:

    python3 ${SKILL_ROOT}/../skill-shaper/scripts/generate_review.py <workspace> --skill-name <name>

  4. Open the HTML viewer in the browser (or use --static <path> for headless environments).

  5. Wait for user feedback from the viewer.