// agents · evalstate/fast-agent

fast-agent

fast-agent é um projeto de agentes de IA de código aberto de evalstate (evalstate/fast-agent no GitHub). Tem 3.928 estrelas no GitHub e 446 forks, é publicado sob a licença Apache-2.0 e foi atualizado pela última vez no GitHub em 8 de outubro de 2026.

Code, build and evaluate LLM agents with full MCP, Skills, ACP and A2A support.

★ 3,9 mil estrelas⑂ 446 forksApache-2.0atualizado out 2026

Ver no GitHub ↗Official site ↗

▦ visual: o README deste projeto não inclui nenhuma captura incorporável,
então aqui vai um marcador. O guia abaixo está completo mesmo assim.

// dados do projeto

Dados do projeto

Repositório: evalstate/fast-agent
Estrelas no GitHub: 3.928
Forks: 446
Licença: Apache-2.0
Última atualização no GitHub: 8 de outubro de 2026
Categoria: Agentes de IA
Tipo de ficha: Guia de instalação completo
Site oficial: Official site

// como funciona

Como funciona

Clarify your goal
Decide the first concrete outcome you want from it, so you can pick the right example, template or mode instead of exploring everything at once.
Get the project
Clone the repository (or install the published package) and skim its README quick start for the requirements for your platform.
Install and configure
Run the documented install step and set any required configuration — such as provider keys, data sources or local services — before the first run.
Run your first task
Execute the smallest documented example end to end, confirm the output is what you expected, then adapt it to your real data or workflow.

// recursos

Por que as pessoas gostam tanto

Complete MCP support

End-to-end tested MCP features including Sampling and Elicitations, plus MCP Apps and the OpenAI Apps SDK.

Agent Skills

SKILL.md-based skills managed with the /skills command, with a built-in registry covering LSP, hooks, compaction and automation.

Built-in workflows

Chain, Parallel, Router, Orchestrator, Evaluator-Optimizer, MAKER and Agents-as-Tools patterns ready to compose.

Interactive shell mode

Run shell commands with ! inside the session and connect to MCP servers with /connect over stdio or streamable HTTP with OAuth.

Multimodal and structured outputs

Vision, PDF and structured-output support, well tested across providers.

ACP and A2A ready

Any setup can run inside ACP clients via fast-agent-acp, with A2A support for agent-to-agent work.

// execute você mesmo

Execute você mesmo

Clone the repository
Get the source so you can inspect, configure and run it locally.
```
git clone https://github.com/evalstate/fast-agent.git
cd fast-agent
```
Install fast-agent
Install the package as a uv tool so the CLI and shell aliases are available.
```
uv tool install -U fast-agent-mcp
```
Start an interactive session
Launch with shell support and a model of your choice, then connect MCP servers with /connect.
```
fast-agent --model opus -xx  # or quick-try: uvx fast-agent-mcp@latest -x
```

// casos de uso

O que você pode criar

Build a support agent that uses MCP tools to answer account questions
Evaluate a multi-step research workflow before shipping it
Use it as a coding agent with Hugging Face or Codex packs
Prototype an agent idea in minutes, then turn it into a Python app

Readers working through the AI Agents shelf also stop at Guardrails, Programmable guardrails to keep LLM systems safe and on rails. Also on this shelf: council-of-high-intelligence — Convene 18 legendary thinkers as an AI council that debates your hardest questions — inside Claude Code. Readers working through the AI Agents shelf also stop at AP2, an open protocol for secure AI-driven payments.

// perguntas frequentes

Perguntas frequentes

Who maintains fast-agent?

It is maintained by evalstate on GitHub (evalstate/fast-agent), where the source, issues and releases live.

Is it free to use?

Yes — it is free and open source under Apache-2.0, which permits use, modification and commercial redistribution, with licence/notice retention.

Does it need a paid API key?

Yes — most use needs your own LLM provider key (Anthropic, OpenAI, Google or another supported provider), unless you point it at a local model via Ollama or llama.cpp.

How does it compare to LangChain or the OpenAI Agents SDK?

Compared with LangChain or the OpenAI Agents SDK, fast-agent's distinguishing focus is: code, build and evaluate LLM agents with full MCP, Skills, ACP and A2A support. Pick it when that focus matches your need; the alternative may suit you better if its broader or different scope is the priority.

How do I get started?

Os outros projetos de Agentes de IA com mais estrelas no catálogo do LibreAI — a lista realista de alternativas se fast-agent não for o ideal para você.

openclaw★ 391,6 mil
A personal AI that really does things — on any OS, with your data staying yours.
superpowers★ 294,9 mil
An agentic skills framework & software development methodology that works.
ECC★ 275,2 mil
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
hermes-agent★ 252,0 mil
The agent that grows with you — a Python framework for adaptive AI agents.
deepseek-harness★ 245,5 mil
DeepSeek's plugin-based agent harness where everything is a plugin.

// mais em agents

Mais em Agentes de IA

Projetos de Agentes de IA próximos no catálogo, os mais próximos em estrelas de fast-agent: continue explorando dos dois lados dele.

cumora★ 3,9 mil
Team chat where AI agents are first-class teammates, not bots.
notfair-plugin★ 3,9 mil
SEO, GEO, and marketing skills your agents can use.
playwriter★ 4,0 mil
Playwriter lets AI agents drive your real Chrome — your logins, extensions, and cookies — via Playwright
openai-agents-js★ 3,9 mil
OpenAI's lightweight TypeScript framework for multi-agent workflows and voice agents.
OpenScience★ 4,0 mil
OpenScience — the open-source AI workbench for scientific research: give it a goal, it reads the literature, runs the experiments and writes up the findings
LazyLLM★ 3,9 mil
A low-code tool for building multi-agent LLM apps — chatbots, RAG and fine-tuning in one flow.

Baixe o fast-agent

Grátis e aberto: pegue o código-fonte, dê uma estrela no repositório e comece a criar.

Ver no GitHub ↗← Mais ferramentas de Agentes de IA