skill-authoring

v0.1.0

Author and eval agent skills: write-skills (triggers, leading words, progressive disclosure, failure modes to prune) and eval-skills (blind golden-case runs in fresh subagents, judged against a bar, gap-driven edits). Vendored from dzhng/skills (MIT).

CodexClaude Code2 skills

by Aviv SinaiMIT0updated 1 month ago

Source

$ git clone https://github.com/avivsinai/skill-authoring

Clone the source, then follow the repository's marketplace instructions for your runtime. The repository root is the plugin root.

Layout

skill-authoring/

├── .codex-plugin/plugin.json
├── .claude-plugin/plugin.json
├── skills/eval-skills/SKILL.md
└── skills/write-skills/SKILL.md

Skills2

eval-skillsskills/eval-skills/SKILL.md

Eval and improve a skill against golden cases — run the target skill blind in a fresh, context-free subagent on each example input, grade the artifact against a per-case bar, and let the gaps drive the edits. Use when the user wants to test/eval/improve/harden a skill, says "this skill keeps producing X / keeps missing Y", or hands a skill plus cases with a bar for what good looks like. Pairs with write-skills (the authoring principles every fix obeys). Judgment-first and harness-free — prefer skill-creator for harness-based benchmarking, variance analysis, or description-optimization loops.

write-skillsskills/write-skills/SKILL.md

Create or revise agent skills. Use when adding a new skill file, renaming a skill, simplifying an existing skill, improving trigger descriptions, or deciding what belongs in a skill versus references, scripts, assets, or ordinary docs. This is the judgment layer — authoring principles and failure modes for manual, principle-level work; prefer skill-creator when the user wants end-to-end interactive creation with generated test harnesses or quantitative description optimization.

Manifests2

.codex-plugin/plugin.json

{
  "name": "skill-authoring",
  "version": "0.1.0",
  "description": "Author and eval agent skills: write-skills (triggers, leading words, progressive disclosure, failure modes to prune) and eval-skills (blind golden-case runs in fresh subagents, judged against a bar, gap-driven edits). Vendored from dzhng/skills (MIT).",
  "author": {
    "name": "Aviv Sinai",
    "url": "https://github.com/avivsinai"
  },
  "repository": "https://github.com/avivsinai/skill-authoring",
  "license": "MIT",
  "skills": "./skills/",
  "keywords": [
    "skills",
    "authoring",
    "eval",
    "skill-writing",
    "progressive-disclosure",
    "agent-skills",
    "workflow"
  ],
  "interface": {
    "displayName": "Skill Authoring",
    "shortDescription": "Write sharp agent skills and eval them blind against golden cases",
    "category": "Developer Tools"
  }
}

.claude-plugin/plugin.json

{
  "name": "skill-authoring",
  "version": "0.1.0",
  "description": "Author and eval agent skills: write-skills (triggers, leading words, progressive disclosure, failure modes to prune) and eval-skills (blind golden-case runs in fresh subagents, judged against a bar, gap-driven edits). Vendored from dzhng/skills (MIT).",
  "author": {
    "name": "Aviv Sinai",
    "url": "https://github.com/avivsinai"
  },
  "repository": "https://github.com/avivsinai/skill-authoring",
  "license": "MIT",
  "skills": "./skills/",
  "keywords": [
    "skills",
    "authoring",
    "eval",
    "skill-writing",
    "progressive-disclosure",
    "agent-skills",
    "workflow"
  ]
}