EdgeOne Makers

Official AI Agent Skills for developing and deploying projects on EdgeOne Makers.

Installation

Option A — npx skills (Claude Code / Cursor / Codex / CodeBuddy CLI 等)

npx skills add TencentEdgeOne/edgeone-makers-tools

This installs 9 independent sub-skills (one per capability) into your agent's skills directory (.codebuddy/skills/<name>/, .claude/skills/<name>/, .cursor/skills/<name>/, …). Your agent loads only the matching skill per task — no router indirection.

Option B — CodeBuddy plugin marketplace / SkillHub

Search and install edgeone-makers-tools from the CodeBuddy plugin marketplace or SkillHub. These platforms install the skill from the skillhub branch, which adds a frontmatter to the root SKILL.md so the platform can register it as a single root skill (it then routes to the same 9 sub-skills internally).

Why two branches? The two install ecosystems have incompatible requirements on the root SKILL.md (one wants no frontmatter, the other requires one). The main branch is optimized for npx skills add; the skillhub branch is optimized for CodeBuddy / SkillHub. See BRANCH.md on the skillhub branch for the maintenance flow.

Option C — Claude Code plugin marketplace

/plugin marketplace add TencentEdgeOne/edgeone-makers-tools
/plugin install edgeone-makers-tools@edgeone-makers

After installation, your AI coding agent will automatically detect relevant tasks and load the right skill.

Skills

Skill	Description
makers-agents	AI Agent development (DeepAgents, LangGraph, Claude SDK, OpenAI Agents, CrewAI)
makers-deploy	Deploy projects to EdgeOne
makers-edge-functions	Edge Functions (V8 lightweight runtime)
makers-cloud-functions	Cloud Functions (Node.js / Go / Python)
makers-storage	KV + Blob Storage
makers-middleware	Middleware (auth, rewrites, routing)
makers-cli	CLI command reference
makers-recipes	Project structure templates & scaffolding

Usage Examples

Deployment:

Deploy my project to EdgeOne

Deploy this Next.js project and give me the preview URL

Development:

Create an API for user registration

Write middleware to protect my /api routes with auth

Set up Edge Functions with KV storage for a page view counter

Create a Go API with Gin framework

AI Agents:

Build an AI chat agent on EdgeOne Makers

Wire LangGraph into my Makers project with checkpointer + store

Create a Claude Agent SDK endpoint with sandbox code execution

Skill Structure

skills/
├── makers-agents/               # AI Agent development
│   ├── SKILL.md                 # Decision tree, red lines, framework routing
│   └── references/
│       ├── platform/            # Entry conventions, env, SSE protocol
│       ├── capabilities/        # Store, sandbox, tools
│       ├── node-frameworks/     # Claude SDK, LangGraph, OpenAI Agents, DeepAgents
│       └── python-frameworks/   # Claude SDK, LangGraph, OpenAI Agents, DeepAgents, CrewAI
├── makers-deploy/               # Deployment workflow
│   ├── SKILL.md
│   └── references/
├── makers-edge-functions/       # V8 edge runtime
│   └── SKILL.md
├── makers-cloud-functions/      # Node.js / Go / Python
│   ├── SKILL.md
│   └── references/
├── makers-storage/              # KV + Blob storage
│   └── SKILL.md
├── makers-middleware/           # Request interception
│   └── SKILL.md
├── makers-cli/                  # CLI commands
│   └── SKILL.md
└── makers-recipes/              # Project templates
    └── SKILL.md

Each skill follows the skill-creator standard:

• SKILL.md — YAML frontmatter (name + description) + core instructions
• references/ — detailed docs loaded on demand, routed from SKILL.md

Multi-Platform Support

This repo includes plugin manifests for multiple AI platforms:

• .claude-plugin/ — Claude Code
• .cursor-plugin/ — Cursor
• .codebuddy-plugin/ — CodeBuddy

Hooks (hooks/) provide context-aware skill injection via UserPromptSubmit and PreToolUse events.

Requirements

• Node.js ≥ 16
• An EdgeOne account: China site | Global site

License

MIT