#8787th of 142 in Agent frameworks and SDKs
fast-agent
Python framework and CLI for building MCP-enabled agents and workflows
- Stars
- 4k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- Python
3,707 on 22 Apr 20263,965 on 11 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
fast-agent is a Python framework and CLI for defining agents with decorators and connecting them to MCP servers. It provides chain, parallel, router, orchestrator, evaluator-optimizer and MAKER workflows. It also runs as an interactive shell-enabled coding agent or as an ACP-compatible agent. Native providers are Anthropic, OpenAI and Google, plus Azure, Ollama and others via TensorZero.
Who it is for: Python developers building MCP-connected agents and workflows from the terminal
Strengths
- Covers MCP features: sampling, elicitations, OAuth with keyring storage, Streamable HTTP diagnostics
- Agents defined with short decorators; prompts and config live in plain files for version control
- Built-in workflows: chain, parallel, router, orchestrator, evaluator-optimizer, MAKER, agents-as-tools
- Passthrough and Playback LLMs let you test glue code without calling a real model
Weaknesses
- CLI-first; no web UI or visual workflow builder
- Python only; setup instructions assume the uv package manager
- No Docker or compose files provided
- Providers beyond the native ones depend on TensorZero as an intermediary
What it needs
- no GPU
- Needs uv, MCP servers, LLM provider API or local model server (Ollama, llama.cpp)
- Models: Anthropic, OpenAI, Google, Azure, Ollama
What changed
- v0.10.53 ↗ (opens in a new tab)Breaking
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Agent frameworks and SDKs
See all 142| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 96 out of 100 |
| 2 | openai-agents-pythonPython SDK for multi-agent workflows with handoffs, guardrails and tracing | 92 out of 100 |
| 3 | mastraTypeScript framework for building AI agents, workflows and MCP servers | 92 out of 100 |
| #4 | browser-usePython library that lets an LLM agent control a web browser | 90 out of 100 |
| #5 | copilotkitReact and Angular SDK for adding agent chat and generative UI to apps | 90 out of 100 |