Overview
AgentGPT is an open-source project from Reworkd that lets you build and run autonomous AI agents straight from your web browser. You give an agent a name and a goal, and it tries to reach that goal on its own by coming up with tasks, running them, and learning from what it finds.
Under the hood it pairs a Next.js frontend with a FastAPI backend and a MySQL database, and uses LangChain to drive the language model. You can try the hosted version at agentgpt.reworkd.ai, or clone the repository and run the full stack locally with the bundled setup script.
What it does
- Create custom autonomous agents by giving each one a name and a goal
- Agents break a goal into tasks, execute them, and learn from the results in a loop
- Runs entirely in the browser through a Next.js web interface
- FastAPI backend with LangChain for the agent reasoning and tooling
- Optional Serper API key for web search and Replicate token for extra model access
- Self-hostable full stack (frontend, backend, MySQL) set up by an automatic CLI
Getting started
The easiest way to run AgentGPT locally is the automatic setup CLI bundled with the project. You will need Node.js, Git, Docker, and an OpenAI API key installed first. Serper and Replicate keys are optional.
Clone the repository
Open a terminal and clone the AgentGPT repository, then move into the project directory.
git clone https://github.com/reworkd/AgentGPT.git
cd AgentGPTRun the setup script
Run the bundled setup script for your operating system. It configures the environment variables, database, backend, and frontend for you.
# Mac/Linux
./setup.sh
# Windows
./setup.batAdd your keys and open the app
Follow the prompts from the setup script to add the appropriate API keys. Once all services are running, open http://localhost:3000 in your browser to start creating agents.
Commands and code are distilled from the project's own documentation — always check the official repo for the latest.
When to use it
- Spin up an autonomous agent that researches a topic and reports back its findings
- Prototype goal-driven AI workflows without writing a full agent loop from scratch
- Self-host a private agent platform for your team using the included Docker setup
- Experiment with how an LLM plans, executes, and refines tasks toward a single goal
How AgentGPT compares
AgentGPT alongside other open-source agent frameworks & builders tools AI/TLDR tracks, ranked by GitHub stars.
| Tool | Stars | What it does |
|---|---|---|
| AutoGPT | ★ 185k | One of the earliest autonomous agent projects, now a platform for building and running agents from reusable blocks and workflows. |
| Agno | ★ 40.8k | A fast Python framework (formerly Phidata) for building agents with memory, tools, and multimodal inputs, plus a runtime for deploying them in production. |
| AgentGPT | ★ 36.2k | Assemble, configure, and deploy autonomous AI agents in your browser |
| LangGraph | ★ 35.2k | A library from the LangChain team for building stateful, graph-based agent workflows with explicit control over steps, memory, and human-in-the-loop checkpoints. |
| Composio | ★ 28.9k | Composio is an open-source SDK for Python and TypeScript that gives AI agents ready-made tools to act on real apps and APIs across many agent frameworks. |
| Semantic Kernel | ★ 28.2k | Microsoft's SDK for adding agents, plugins, and planning to apps across .NET, Python, and Java. |
| smolagents | ★ 27.9k | A minimal agent library from Hugging Face where the model writes and runs Python code to call tools and complete tasks. |
| Mastra | ★ 25.3k | A TypeScript framework for building AI agents and applications with workflows, RAG, memory, and observability built in. |