Claude Code skill for building production AI agents with the Claude Agent SDK. Covers Python and TypeScript patterns for query(), ClaudeSDKClient, custom tools, hooks, subagents, and multi-agent orchestration.
A Claude Code skill for building production AI agents with the Claude Agent SDK.
Plugin (recommended):
/install-plugin github:tkwn2080/claude-agent-sdk-skill
Manual:
git clone https://github.com/tkwn2080/claude-agent-sdk-skill.git
cp -r claude-agent-sdk-skill/skills/claude-agent-sdk ~/.claude/skills/
Restart Claude Code after installing.
When you work with the Agent SDK, Claude automatically applies correct patterns for:
query() and ClaudeSDKClient usagemcp__{server}__{tool} naming, streaming input)| Topic | File | What's covered |
|---|---|---|
| Core APIs | SKILL.md | query(), ClaudeSDKClient, configuration options, authentication (incl. OpenRouter) |
| Python | references/python-patterns.md | All patterns: streaming input, hooks, permissions, can_use_tool, error handling |
| TypeScript | references/typescript-patterns.md | All patterns: Zod schemas, hooks, canUseTool, Query object methods |
| Hooks | references/hooks-reference.md | All 12 hook events, permission evaluation order, acceptEdits scope, troubleshooting |
| Custom tools | references/custom-tools.md | @tool decorator, createSdkMcpServer, external MCP servers, annotations |
| Multi-agent | references/multi-agent.md | Orchestrator-worker, routing, parallelization, evaluator-optimizer, prompt chaining |
| Known issues | Across all files | Stream closed fixes, MCP + subagent limitations, can_use_tool bugs, headless mode |
skills/claude-agent-sdk/
├── SKILL.md # Entry point — quick start, config, known issues
└── references/
├── python-patterns.md # Python SDK patterns and examples
├── typescript-patterns.md # TypeScript SDK patterns and examples
├── hooks-reference.md # Hooks, permissions, troubleshooting
├── custom-tools.md # Tool creation and MCP servers
└── multi-agent.md # Multi-agent orchestration patterns
MIT
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub ettrickshepherd/claude-agent-sdk-skillClaude Code skill for writing idiomatic Apple MLX code on Apple Silicon. Covers lazy evaluation, array indexing, neural networks, compilation, memory management, and PyTorch migration.
Claude Agent SDK Development Plugin
Anthropic Claude Agent SDK development for Python and TypeScript autonomous agents
Anthropic Claude Agent SDK for autonomous agents and multi-step workflows. Use for subagents, tool orchestration, MCP servers, or encountering CLI not found, context length exceeded errors.
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques