Implement skill registry structure with auto-generated index and CI validation

[Copilot]

Establishes the foundational structure for a centralized skill registry with automated indexing and validation.

Structure

• skills/ - Registry root with auto-generated index.json catalog
• scripts/generate-index.js - Scans skill directories, validates metadata, generates sorted index
• .github/workflows/validate.yml - CI validates JSON schema, required fields, and index freshness
• README.md - Usage patterns for adding/discovering skills

Skill Format

Each skill lives in skills/<skill-name>/ with:

• skill.json - Metadata (name, version, description, author, tags, repository)
• README.md - Documentation

Example:

{
  "name": "example-skill",
  "version": "1.0.0",
  "description": "An example skill demonstrating the skill structure",
  "author": "Refly AI Team",
  "tags": ["example", "demo"],
  "repository": "https://github.com/refly-ai/refly-skills"
}

Validation

Workflow enforces:

• Valid JSON in all skill.json files
• Required name field presence
• Index synchronization with skill directories
• Read-only GITHUB_TOKEN permissions

Included example-skill/ demonstrates the pattern.

Original prompt

refly-ai/skill-registry/
├── README.md # 仓库介绍、如何使用
├── skills/
│ ├── index.json # 所有 skills 的索引（自动生成）
│ └── /
│ ├── skill.json # 元数据
│ └── README.md # 自动生成的文档
├── .github/
│ └── workflows/
│ └── validate.yml # PR 验证（格式检查）
└── scripts/
└── generate-index.js # 生成 index.json 的脚本

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[alchemistklk]

Approved by automation