$ recombobulate _
home / tips / build-a-custom-agent-with-the-claude-agent-sdk-in-10-lines
30

Build a Custom Agent with the Claude Agent SDK in 10 Lines

recombobulate @recombobulate · Mar 26, 2026 · Workflows
build-a-custom-agent-with-the-claude-agent-sdk-in-10-lines

The Claude Agent SDK (formerly Claude Code SDK) gives you the same tools and agent loop that power Claude Code, but as a library you can call from Python or TypeScript. No tool implementations needed.

import asyncio
from claude_agent_sdk import query, ClaudeAgentOptions

async def main():
    async for message in query(
        prompt="Find all TODO comments and create a summary",
        options=ClaudeAgentOptions(
            allowed_tools=["Read", "Glob", "Grep"]
        ),
    ):
        if hasattr(message, "result"):
            print(message.result)

asyncio.run(main())

Built-in tools include Read, Write, Edit, Bash, Glob, Grep, WebSearch, and WebFetch. You don't have to implement any of them. Just list what the agent should have access to in allowed_tools.

Install with pip install claude-agent-sdk or npm install @anthropic-ai/claude-agent-sdk, set ANTHROPIC_API_KEY, and you're running. The SDK also supports Bedrock, Vertex AI, and Azure for enterprise auth.

Sessions persist automatically. Capture session_id from the init message and pass it back with resume=session_id to continue a conversation with full context.

Same power as Claude Code, now programmable in your own applications.


via Claude Code Docs — SDK

~/recombobulate $ tip --comments --count=0

Log in to leave a comment.

~/recombobulate $ tip --related --limit=3
0
Run Claude Code in GitHub Actions to Automatically Review Every Pull Request

Set up Claude Code as an automated reviewer in your CI pipeline — on every pull request, it reads the diff, checks for bugs, security issues, missing tests, and convention violations, then posts its findings as a PR comment. Your human reviewers get a head start because the obvious issues are already flagged before they look.

recombobulate @recombobulate · 1 month ago
0
Ask Claude to Build a Deployment Checklist from Your Actual Infrastructure

Before deploying, tell Claude to read your project — migrations, environment variables, queue workers, scheduled tasks, caching, third-party integrations — and generate a deployment checklist that's specific to your app. Not a generic "did you run migrations?" list, but one that knows YOUR infrastructure and catches the things YOUR deploy can break.

recombobulate @recombobulate · 1 month ago
0
Ask Claude to Generate a README from Your Actual Codebase — Not a Template

Instead of writing a README from memory or copying a template, tell Claude to read your project and generate one that's actually accurate — real setup instructions from your config, real architecture from your directory structure, real API examples from your routes, and real prerequisites from your dependency files.

recombobulate @recombobulate · 1 month ago