Help us improve
Share bugs, ideas, or general feedback.
From armory
Builds AI agents and automates Claude Code using Agent SDK and headless CLI mode. Covers Python SDK, MCP servers, hooks, sessions, and programmatic queries.
How this skill is triggered — by the user, by Claude, or both
Slash command
/armory:agent-builderThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
**Triggers:**
Share bugs, ideas, or general feedback.
Triggers:
This skill covers Claude Code's headless mode and the Agent SDK — not the raw Anthropic API.
Preferred: Claude Code CLI headless mode (claude -p)
Preferred: Claude Agent SDK (wraps Claude Code CLI)
Avoid: Raw Anthropic API (anthropic.Anthropic())
Avoid: Direct API calls with API keys
Why this matters:
~/.claude/ config — no API keys neededRule: If code uses anthropic.Anthropic(), anthropic.messages.create(), or requires ANTHROPIC_API_KEY, that is the wrong approach. Use claude -p or claude_agent_sdk instead.
claude -p)All use Claude Code's existing authentication from ~/.claude/ — no API keys required.
Load only the reference file relevant to the current task.
| Task | Load |
|---|---|
| Distinguishing raw Anthropic API from Claude Code SDK patterns | references/api-vs-sdk.md |
| Writing a Python agent with the Agent SDK (install, query, client, tools, hooks, errors) | references/sdk-python.md |
Using headless claude -p mode (basic, JSON schema, streaming, tool approval, sessions, system prompts) | references/headless-cli.md |
| CLI commands, flags table, dynamic subagents, permission rule syntax | references/cli-reference.md |
| Use-case templates: CI/CD pipeline, code review, custom analysis tool, database query agent | references/use-cases.md |
| Agent SDK and CLI/Headless best practices | references/best-practices.md |
| Complete end-to-end agent example (tools + hooks + client) | references/example-agent.md |
| Output format conventions for produced artifacts | references/output-format.md |
--dangerously-skip-permissions in production--max-budget-usdNeed to...
claude -p "query"ClaudeSDKClient@tool decoratorHookMatcher--output-format stream-json--json-schema--continue or --resumeclaude -p with --allowedToolsquery() or ClaudeSDKClientFetch live documentation when:
Use WebFetch tool to pull latest documentation from source URLs.
Before writing code:
claude -p or claude_agent_sdk) — not raw Anthropic API~/.claude/ configwhich claude)uv pip listAfter writing code:
claude_agent_sdk or claude -p — not anthropic.Anthropic()ANTHROPIC_API_KEY or api_key= parameters in codeThis skill provides:
Ask specific questions about building agents, running headless commands, or CLI usage.
~/.claude/ before any SDK or headless usage works.anthropic.Anthropic()); see the Anthropic API docs for that.--resume or session ID capture.npx claudepluginhub mathews-tom/armory --plugin armoryBuilds autonomous AI agents with Claude Agent SDK: computer use, tool calling, MCP integration, and production best practices for Anthropic models.
Builds and troubleshoots Claude Agent SDK apps in Python and TypeScript, covering APIs, sessions, permissions, streaming, tools, plugins, and extensibility.
Implements Anthropic Claude Agent SDK for autonomous agents, subagents, tool orchestration, MCP servers, and multi-step workflows. Useful for session management, permissions, and errors like CLI not found or context exceeded.