Update to Plugin Eval

{
  "name": "improve-skill",
  "description": "Turn plugin-eval findings into a concrete rewrite brief for a Codex skill. Use when the user already evaluated a skill and now wants Codex to improve it, especially after asking what to fix first.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 113
    }
  ],
  "skill_md_contents": "---\nname: improve-skill\ndescription: Turn plugin-eval findings into a concrete rewrite brief for a Codex skill. Use when the user already evaluated a skill and now wants Codex to improve it, especially after asking what to fix first.\n---\n\n# Improve Skill\n\nUse this skill after `plugin-eval` has already produced findings for a local skill.\n\n## Workflow\n\n1. Run `plugin-eval analyze <skill-path> --brief-out <brief.json>`.\n2. Read the improvement brief and group work into required fixes versus recommended fixes.\n3. Apply the `skill-creator` guidance from `/Users/benlesh/.codex/skills/skill-creator/SKILL.md`.\n4. Re-run the evaluation and compare before and after outputs.\n\n## Chat Requests To Recognize\n\n- `Improve this skill based on the evaluation.`\n- `Rewrite this skill using the plugin-eval findings.`\n- `What should I fix first in this skill?`\n\n## Focus Areas\n\n- reduce trigger and invoke token costs\n- keep `SKILL.md` compact\n- move bulky details into references or scripts\n- improve trigger descriptions\n- fix broken links and manifest/frontmatter issues\n\n## Commands\n\n```bash\nplugin-eval analyze <skill-path> --brief-out ./skill-brief.json\nplugin-eval compare before.json after.json\n```\n\n## Reference\n\n- `../../references/chat-first-workflows.md`\n"
}