Claude Agent SDK

Want to build an agent with the same engine behind Claude Code? The Claude Agent SDK is a developer toolkit from Anthropic. In Anthropic's words, it gives you the same tools, agent loop, and context management that power Claude Code, as a library you can program in Python and TypeScript (we checked the docs on Oct 28, 2026).

If you've seen our Claude Code profile, think of it this way. Claude Code is the ready-to-use tool. The Agent SDK lets developers put that same engine inside their own app.

Anthropic compares four ways to build:

• Agent SDK: embed Claude Code's agent in your own Python or TypeScript app.
• Claude Code CLI: the terminal tool, for day-to-day use.
• Client SDK: call the Claude API directly and write the loop yourself.
• Managed Agents: Anthropic hosts the agent for you, in its cloud or in your own sandbox.

Here's what comes built in, per the docs:

• Tools to read, write, and edit files, run commands, and search the web.
• Permissions: you choose which tools run on their own and which need approval.
• Subagents for focused side tasks.
• MCP, a standard way to plug in outside tools and data.
• Sessions that keep context, and can be resumed or forked later.
• Hooks, skills, memory, and plugins.

A couple of rules worth knowing if you build a product with it. Anthropic says you should use API key sign-in, not claude.ai logins, unless it's approved you otherwise. And your product needs its own name and look. You can say it's "Powered by Claude," but you can't call it Claude Code. Use is governed by Anthropic's Commercial Terms of Service.

Pricing: Anthropic's site has current pricing. You pay for Claude API usage, and those rates aren't on this page, so pricing is unknown here for now.