Log inSubmit server
Metorial logo

Metorial

by metorial

Connects AI agents to external systems through managed integrations, authentication, permissions, and auditability.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Metorial

Pick your client, then add it in one click or paste the config.

STDIOPython
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "metorial-metorial": {
      "command": "uvx",
      "args": [
        "metorial"
      ],
      "env": {
        "METORIAL_API_KEY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

Install not yet sandbox-tested— we haven't run this command through our automated sandbox yet. Not a red flag; we're working through the catalog.

Overview

The metorial/metorial MCP server connects AI agents to external systems through Metorial’s integration catalog and unified access layer. It manages OAuth, API keys, service accounts, scoped permissions, and audit logs while exposing connected integrations as agent tools. Use it when multiple agents or projects need consistent access to SaaS tools, enterprise systems, data sources, or custom MCP servers. The repository also provides JavaScript/TypeScript and Python SDK paths for connecting sessions to agent frameworks.

Use cases

  • Connect agents to SaaS and enterprise systems
  • Centralize OAuth and credential handling
  • Apply scoped permissions to agent workflows
  • Audit actions performed by connected agents
  • Reuse provider connections across projects

Key features

  • 1200+ integrations
  • OAuth and token lifecycle management
  • RBAC, SAML SSO, and IAM
  • Scoped agent permissions
  • Audit logs and action traceability
  • JavaScript and Python SDKs

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked Today

Recent health check succeeded.

31/31 checks healthy over the last 78d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 10d ago via OSV.dev · metorial (PyPI)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars3.4kStargazers on the source repository.
Last commitTodayMost recent push to the default branch.
Directory activity1 viewsConfig copies, upvotes, and views on AllMCPs.

How Metorial works

What metorial/metorial MCP server does

The metorial/metorial MCP server provides a control plane between AI agents and external systems. It gives agents access to integrations spanning SaaS applications, enterprise platforms, data sources, and custom MCP servers through a common interface. The repository describes more than 1,200 integrations, including examples such as Slack, GitHub, Google Calendar, and SAP.

Rather than implementing each integration and credential flow separately, applications configure providers through Metorial. Connected providers can then be exposed as tools to an agent framework. This is relevant for teams that need shared access rules, reusable connections, or visibility into agent activity across projects and environments.

How it works

Applications initialize the Metorial client with an API key, create or select a provider deployment, and open a connection using an agent-framework adapter. The resulting session exposes its available tools through session.tools(), which can be passed to an AI SDK or agent.

Authentication can use OAuth, API keys, service accounts, and other supported credential flows. For user-authorized services, the application creates a setup session and presents its URL to the user. After the user completes authentication, the returned auth configuration is supplied when establishing the provider connection.

Provider configuration can combine dashboard-managed deployments, existing authentication configurations, inline credentials, and reusable session templates. The platform applies access controls and records agent actions, including the agent involved and the credentials used.

Setup and configuration

The documented quick start requires a METORIAL_API_KEY. JavaScript and TypeScript applications install metorial, @metorial/ai-sdk, @ai-sdk/anthropic, and ai; Python applications install metorial, pydantic-ai, and python-dotenv for the shown example.

The JavaScript path uses the Metorial client with the AI SDK adapter. The Python path uses the Metorial client with the Pydantic AI adapter. Both examples create a Metorial Search deployment, connect it to an agent session, and pass the session tools to a model.

OAuth setup additionally requires provider and authentication-method identifiers. The Python example accepts a redirect URL for the OAuth callback. After setup completes, the resulting authentication configuration can be reused by later connections.

The Metorial Platform repository contains the engine and can be self-hosted. The README also points to a hosted platform for avoiding local platform setup.

Tools and capabilities

The metorial/metorial MCP server supports these documented capabilities:

  • Exposing configured integrations as tools for agent frameworks.
  • Connecting to a catalog of SaaS, enterprise, data, and custom MCP providers.
  • Managing OAuth and other credential lifecycles.
  • Applying RBAC, SAML SSO, IAM, and scoped permissions.
  • Reusing connections across projects, environments, teams, and people.
  • Recording audit logs and agent-action traceability.
  • Providing a built-in Metorial Search provider that requires no authentication configuration.

Limitations and notes

Provider-specific setup may require authentication-method and deployment identifiers.

The examples depend on external agent frameworks: the JavaScript example uses the AI SDK and Anthropic provider, while the Python example uses Pydantic AI. Self-hosting refers to the separate Metorial Platform repository rather than only this repository.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Metorial

metorial/metorial is an MCP server that connects AI agents to SaaS tools, enterprise systems, data sources, and custom MCP servers through Metorial’s integration layer. Its main tools expose configured providers to agent frameworks, while its access layer manages OAuth, API keys, permissions, reusable connections, and audit visibility.

Install the documented SDK path for your language: use npm packages for JavaScript or TypeScript, or `pip install metorial pydantic-ai python-dotenv` for Python. The examples then initialize the client with `METORIAL_API_KEY` and connect a provider to an agent framework.

Create a provider setup session, send its URL to the user for browser authentication, wait for completion, and pass the returned authentication configuration when connecting the provider. The setup requires the provider and authentication-method identifiers.

The README names Cursor as a supported environment for using Metorial integrations.

Related MCP servers

All in Other Tools and Integrations All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewMetorial AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/metorial-metorial?style=directory)](https://allmcps.com/mcp/metorial-metorial)
HTML Embed
<a href="https://allmcps.com/mcp/metorial-metorial"><img src="https://allmcps.com/api/badge/metorial-metorial?style=directory" alt="Metorial on AllMCPs" /></a>