Arcade MCP

A Python framework and CLI for building MCP servers and tools with built-in OAuth authorization

CLI

Medium risk

We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.

Why this level

  • Builds MCP servers that access external APIs
  • Handles OAuth tokens through the platform
All reasons and checks

arcadeai/arcade-mcp

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add cli/arcade-mcp

Detects the agents on your machine, checks the risk and pins the version.

Other ways to install

Install the tool

pipx install arcade-mcp

Without pipx, pip install arcade-mcp works too.

You will need: pipx

Checked against the repository on Sep 25, 2026, commit e0f09df.

Text for your agent

Install the CLI with uv tool install arcade-mcp. Then use the decorators to describe tools and the arcade commands for tests and deployment.

Other ways from the author
uv tool install arcade-mcp

Installs the arcade CLI; a source build is available via git clone.

This is third-party code. Review the repository files before installing.

What it does

Arcade MCP is a Python framework for building your own MCP servers and tools when the prebuilt catalog lacks one: internal APIs, custom OAuth providers, domain integrations. Tools are described with decorators covering the MCP spec: tools, resources, prompts, sampling and logging. A distinct feature is authorized tool calling: a tool declares the OAuth scopes it needs, and the platform handles the user login, token storage and refresh, while the client and the model never see the token. There is an arcade CLI with commands to test call accuracy and to deploy.

Who it is for. For developers who build their own MCP servers and tools for agents.

Good fit when

  • You need to write your own MCP server for an internal API
  • You want built-in OAuth authorization for tools
  • You want to test tool-call accuracy with a model

Not a fit when

  • An existing MCP server already covers the task
  • You do not need to write your own tools

Example request

Help me write an MCP tool for our internal API with OAuth authorization

Limitations

It is developer tooling rather than a ready server for a specific service. Token management and hosting of some features tie into the Arcade platform, and cloud deployment via arcade deploy is optional. It requires Python and uv.

How to disable. Remove the CLI with uv tool uninstall arcade-mcp and drop the package from the project's dependencies.

Security check

  • Builds MCP servers that access external APIs
  • Handles OAuth tokens through the platform

README in short

The README presents arcade-mcp as an open-source Python framework for building MCP servers and tools, and the engine behind Arcade's large prebuilt tool catalog. Tools are described with decorators across the MCP spec, and authorized calling with OAuth and platform-side token storage is highlighted. The CLI installs with uv tool install arcade-mcp, and there are arcade evals and arcade deploy commands. MIT licensed.

FAQ

Is it a ready MCP server?

No, it is a framework and CLI for building your own servers and tools when prebuilt ones fall short.

How does tool authorization work?

A tool declares the OAuth scopes it needs, the platform handles login and tokens, and the client and model never see the token.

Editors’ pick

Open-source personal AI assistant on your own machine: answers in Telegram, Slack, Discord and WhatsApp, extended with skills and plugins

CLIHigh risk390.7KRepository stars
Editors’ pick

A self-improving agent from Nous Research with a TUI, messaging gateway, cron jobs and skills it writes itself

CLIHigh risk249.8KRepository stars
Editors’ pick

An open source coding agent for the terminal and desktop with build and plan modes

CLIHigh risk210.6KRepository stars
Editors’ pick

Open prompt library with a Claude Code plugin, MCP server and CLI: search, fetch and improve prompts and skills from an agent

PluginMedium risk171.5KRepository stars
Foxx AIArcade MCP

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.