shipany-ai

shipany-ai / open-agent-sdk

Public

Agent-SDK without CLI dependencies, as an alternative to claude-agent-sdk, completely open source

agent-sdk claude-agent-sdk claude-code open-agent-sdk
399
143
69% credibility
Found Mar 31, 2026 at 399 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Open Agent SDK is an open-source TypeScript library for creating autonomous AI agents that analyze codebases, edit files, execute shell commands, search the web, and run multi-step workflows, fully compatible with Anthropic's Claude Agent SDK API while running entirely in-process.

How It Works

1
🔍 Discover the AI coding helper

You hear about a simple tool that lets smart AI agents read, edit, and improve your code automatically.

2
📦 Add it to your project

Install the package with a quick command so your code can use AI helpers right away.

3
🔑 Connect your AI brain

Link it to an AI service like Claude so the agents can think and respond.

4
🚀 Launch your first agent

Create an agent and ask it to find bugs or refactor code in your project with one simple call.

5
⚙️ Customize with tools

Add special abilities like web search or file commands to make agents even smarter.

✅ Watch agents work magic

Your AI agents now understand your codebase, make changes, and handle complex tasks on their own.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 399 to 399 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is open-agent-sdk?

Open Agent SDK is a TypeScript library for building autonomous AI agents that analyze codebases, edit files, run shell commands, search the web, and handle multi-step workflows. It mirrors Anthropic's claude-agent-sdk API but runs the full agent loop in-process—no CLI subprocess needed—making it deployable in cloud servers, serverless functions, Docker, or CI/CD pipelines. Set your Anthropic or OpenRouter API key and use `query()` for one-shot tasks or `createAgent()` for persistent sessions with tools like Read, Edit, Bash, and subagents.

Why is it gaining traction?

Unlike claude-agent-sdk or LangChain alternatives, it skips CLI dependencies for seamless cloud deployment while matching features like permissions, memory management, context compression, multi-agent teams, and MCP server integration. Developers on agent sdk github and ai agent sdk github love the drop-in `query()` compatibility, custom tool support, and OpenAI/Anthropic model flexibility via OpenRouter. At 399 stars, it's pulling interest as an open ai agent sdk github option for production without vendor lock-in.

Who should use this?

Backend engineers embedding AI code agents in Node.js apps, especially for serverless automation like bug fixing or PR reviews. DevOps teams in CI/CD needing azure ai agent sdk github-style tools without local setup, or TypeScript devs building microsoft agent sdk github-like workflows with Claude models. Ideal if you're evaluating openai agent sdk open source vs. proprietary options.

Verdict

Grab it for flexible agent deployments—docs and examples are solid, with full API reference and 10+ demos. Maturity is early (399 stars, 0.699999988079071% credibility score), so test in non-critical paths first, but it's a practical claude code agent sdk github alternative today.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.