LibreAI / Agentes de IA / mcp-toolbox
// agents · googleapis/mcp-toolbox

mcp-toolbox

mcp-toolbox é um projeto de agentes de IA de código aberto de googleapis (googleapis/mcp-toolbox no GitHub). Tem 16.624 estrelas no GitHub e 1.753 forks, é publicado sob a licença Apache-2.0 e foi atualizado pela última vez no GitHub em 8 de outubro de 2026.

Google's MCP Toolbox for Databases - give agents safe, structured access to your data

★ 16,6 mil estrelas⑂ 1,8 mil forksApache-2.0atualizado out 2026
▦ 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

Estrelas no GitHub
16.624
Forks
1.753
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

  1. Pick a database

    Choose the database your agent needs - prebuilt configs cover Google Cloud and popular open-source databases.

  2. Get Toolbox

    Download the binary for your OS, use the container image, or run it through npx.

  3. Configure sources and tools

    Use a prebuilt toolset with environment variables, or write a tools.yaml defining sources, tools and toolsets.

  4. Connect your MCP client

    Point Claude, an IDE or your app (via the Toolbox SDKs) at the server and run your first query tool.

// recursos

Por que as pessoas gostam tanto

Prebuilt database tools

Instant standard tools for many databases via the --prebuilt flag, down to per-toolset granularity.

Custom tools in YAML

Define your own SQL tools, sources, toolsets, prompts and resources in a configuration file.

Broad database support

Google Cloud databases plus PostgreSQL, MySQL, MongoDB, Redis, Elasticsearch, ClickHouse, Oracle, TiDB and others per the topic list.

Multiple install routes

Run a downloaded binary, a container image, or npx @toolbox-sdk/server.

Client and SDK integration

Connect any MCP client, or embed Toolbox in your application with the Toolbox SDKs.

Built-in UI option

The server can run with a UI flag for interactive exploration of your tools.

Advertisement
// execute você mesmo

Execute você mesmo

  1. Clone the repository

    Get the source code from GitHub and enter the project directory.

    git clone https://github.com/googleapis/mcp-toolbox.git
    cd mcp-toolbox
  2. Run with a config

    Start the server against your tools.yaml (binary or npx route).

    npx @toolbox-sdk/server --config tools.yaml
    # or with the binary: ./toolbox --config "tools.yaml"
  3. Or use a prebuilt toolset

    Point Toolbox at a supported database with the prebuilt flag and the documented environment variables.

    # toolbox --prebuilt=postgres/data - see README Prebuilt Tools Reference for env vars
// casos de uso

O que você pode criar

  • Let a support agent look up order status in Postgres without raw DB credentials
  • Give an analytics agent governed BigQuery tools instead of open SQL access
  • Connect an IDE assistant to a local MySQL dev database through MCP
  • Expose a small, safe set of database operations to an internal chatbot
Advertisement
// perguntas frequentes

Perguntas frequentes

Who maintains this project?

It is maintained by Google (googleapis) on GitHub.

Is it free to use?

Yes - MCP Toolbox is Apache-2.0 licensed and free to use.

Does it need a paid API key?

Toolbox itself needs no key; you provide your database's own credentials/connection settings, and cloud databases use their normal IAM or service-account auth.

How does it compare to alternatives?

Compared with hand-writing an MCP server per database, Toolbox gives you prebuilt, maintained toolsets and a config-driven custom-tool system from Google's team.

How do I get started?

Download the binary or use npx, run it with --prebuilt for your database or a tools.yaml, and connect your MCP client.

// continue explorando

Ferramentas semelhantes

Agentes de IA

brightdata-mcp

Bright Data's MCP server: web search, scraping, structured data and browser automation for agents (75 tools)

★ 2,7 mil estrelasMIT
Ver o guia →
Agentes de IA

mcp-gateway-registry

An enterprise MCP gateway: governed, auditable tool access for agents.

★ 964 estrelasApache-2.0
Ver o guia →
Agentes de IA

open-connector

An open-source auth gateway connecting 1,500+ SaaS providers to AI agents via SDK, CLI, MCP and HTTP.

★ 6,0 mil estrelasApache-2.0
Ver o guia →
Agentes de IA

google_workspace_mcp

Control Gmail, Calendar, Docs, Sheets, Drive and more from AI — a full Google Workspace MCP server.

★ 3,3 mil estrelasMIT
Ver o guia →
Agentes de IA

mcp-shrimp-task-manager

Shrimp Task Manager: an MCP server that turns natural language into structured, dependent dev tasks.

★ 2,1 mil estrelasMIT
Ver o guia →
Agentes de IA

iwe

A markdown knowledge graph with LSP for editors and MCP for agents.

★ 1,7 mil estrelasApache-2.0
Ver o guia →
// alternativas

Alternativas em Agentes de IA

Os outros projetos de Agentes de IA com mais estrelas no catálogo do LibreAI — a lista realista de alternativas se mcp-toolbox 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 mcp-toolbox: continue explorando dos dois lados dele.

  • weaviate★ 16,9 mil

    Vector search plus structured filtering in one database.

  • ai-berkshire★ 16,6 mil

    AI 时代的伯克希尔:基于 Claude Code / Codex 的价值投资研究框架。巴菲特·芒格·段永平·李录四大师方法论 + 多Agent并行研究。| AI-era Berkshire: a value investing research framework built for Claude Code /…

  • openwiki★ 17,0 mil

    A CLI that writes and maintains agent-ready docs for your codebase.

  • memvid★ 16,6 mil

    A serverless, single-file memory layer for AI agents - replace complex RAG pipelines

  • Memori★ 17,1 mil

    Agent-native memory infrastructure - structured, persistent memory for production agents

  • RagaAI-Catalyst★ 16,2 mil

    Python SDK for Agent AI Observability, Monitoring and Evaluation Framework. Includes features like agent, llm and tools tracing, debugging multi-agentic syst…

Baixe o mcp-toolbox

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