{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreig4wvhnjtgnkg4wkf57i7nprgx7h7edma6sr2mp4rzdanpsy35nsa",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mkg5nya7z4k2"
  },
  "path": "/t/template-for-scalable-agent-skill-creation-looking-for-feedback/1379825#post_1",
  "publishedAt": "2026-04-26T17:50:18.000Z",
  "site": "https://community.openai.com",
  "tags": [
    "https://github.com/TechSpokes/skill-base-template"
  ],
  "textContent": "Hi guys,\n\nRecently I’ve been working on several agent skills, and while the process is not exactly hard, I found it is not very scalable if you want to do it properly every time.\n\nSo I made a public GitHub template for this:\n\nhttps://github.com/TechSpokes/skill-base-template\n\nThe idea is simple: put rough source material, notes, examples, etc. into `.intake/`, then let an AI coding agent turn that into a structured skill repo with `SKILL.md`, docs, references, validation scripts, packaging, and release workflow.\n\nBasically, less “let me manually rebuild the same repo structure again” and more “please, dear future robot coworker, follow the recipe.” LoL.\n\nIt is meant for people who want a repeatable path from skill idea → validated repo → releasable skill package.\n\nWould love to hear what you guys think, especially from anyone building reusable skills or internal agent workflows.\n\nAnd yes, if it saves you a bit of time, don’t forget to star it. My GitHub ego is still in beta.",
  "title": "Template for Scalable Agent Skill Creation (Looking for Feedback)"
}