fast-agent
Code, Build and Evaluate agents - excellent Model and Skills/MCP/ACP/A2A Support
fast-agent is a Python-based AI agent development framework that enables developers to code, build, and evaluate sophisticated multimodal agents with comprehensive Model Context Protocol (MCP) support. The framework provides a flexible CLI-first approach for interacting with LLMs, featuring excellent integration with multiple AI providers including Anthropic, OpenAI, Google, Azure, Ollama, and Deepseek. Developers can compose agents using declarative syntax, chain workflows, connect to MCP servers via stdio or HTTP transports, and leverage advanced features like sampling, OAuth authentication, and agent-as-tools patterns. fast-agent stands out as the first framework with complete end-to-end tested MCP feature support including sampling and elicitations, plus unique MCP transport diagnostics for ensuring reliable deployments.
Key Features
Use Cases
- 01Building coding agents with shell support and LSP integration for software development tasks
- 02Creating multi-stage research workflows with web fetching, evaluation, and quality assurance loops
- 03Developing data analysis agents similar to ChatGPT experiences with MCP server integrations
- 04Orchestrating complex tasks by decomposing them into subtasks and routing to specialized sub-agents
- 05Testing and debugging agent-MCP server interactions with different models and transport configurations
- 06Building reliable agent pipelines using MAKER voting to reduce errors in long chains of operations
Related Agents
View morehermes-agent
The agent that grows with you
agency-agents
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
openinterpreter
A coding agent for open models like Kimi K3
cline
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
fast-agent — FAQ
What is fast-agent?+
fast-agent is a Python framework for building and evaluating AI agents with comprehensive Model Context Protocol support. It provides workflows for chaining agents, connecting to MCP servers, and orchestrating complex multi-agent tasks with support for multiple LLM providers.
How do I install fast-agent?+
Install the uv package manager first, then run 'uv pip install fast-agent-mcp' or use 'uv tool install -U fast-agent-mcp' for permanent installation. For quick interactive sessions without installation, use 'uvx fast-agent-mcp@latest -x'.
What AI clients and models does fast-agent work with?+
fast-agent supports Anthropic (Claude), OpenAI (GPT), Google, Azure, Ollama, Deepseek, and dozens of other providers via TensorZero. It can expose agents as MCP servers compatible with any ACP client like Toad using 'fast-agent-acp'.
Do I need API keys to use fast-agent?+
You need API keys for commercial LLM providers like Anthropic, OpenAI, or Google. For local models, you can use Ollama or llama.cpp without API keys by configuring with 'fast-agent model llamacpp' or using the generic provider.
Is fast-agent free to use?+
Yes, fast-agent is open source and free to use under the PyPI license. However, you'll incur costs for commercial LLM API usage based on your chosen providers' pricing.
How do I connect MCP servers to fast-agent?+
Define MCP servers in a 'fast-agent.yaml' configuration file with transport type (stdio or HTTP), command/URL, and optional OAuth settings. In interactive mode, use '/connect @server-name' for stdio servers or '/connect https://url' for HTTP servers.
How do I install fast-agent?+
Open the source repository on GitHub and follow its README. fast-agent is a agent — MCP Agents Market links you directly to the official repo.
Is fast-agent free?+
fast-agent is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.