Appwrite MCP Server

What is this?

The Appwrite MCP Server is a dedicated Model Context Protocol server implementation for Appwrite, an open-source backend-as-a-service platform. It acts as a middleware layer translating MCP protocol calls into Appwrite SDK operations, enabling AI agents and language models to interact with databases, storage buckets, authentication systems, cloud functions, real-time subscriptions, and event triggers seamlessly.

By adopting the MCP standard, it provides a uniform interface for CRUD operations, file uploads, function triggers, and event handling, all secured by Appwrite’s built-in authentication and permission model. This reduces boilerplate, enforces fine-grained access control, and supports private-by-default deployments across public or edge infrastructure, making it ideal for enterprise and research teams building sophisticated agentic workflows.

Quick Start

Install the server using npm:

npm install @modelcontextprotocol/server-appwrite

Then add it to your MCP client configuration:

{
  "mcpServers": {
    "appwrite": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-appwrite"],
      "env": {
        "API_KEY": "your-api-key-here"
      }
    }
  }
}

Key Features

Feature 1: Rapid onboarding—install and configure in minutes using standard NPM commands and environment variables.

Feature 2: Uniform API interface—supports CRUD operations, file management, function triggers, authentication, and real-time data through the MCP protocol.

Feature 3: Secure and extensible workflows—leverages Appwrite’s authentication and permission model for role-based and attribute-based access control, with support for custom functions and webhooks.

Example Usage

One common use case is fetching user purchase history into an LLM context to generate personalized recommendations.

// Example code
const result = await client.callTool({
  name: "documents.list",
  arguments: {
    collectionId: "purchases",
    filters: [["userId", "=", "USER_ID"]],
    limit: 50
  }
});

This code retrieves up to 50 purchase documents for a given userId, which can then be passed to an LLM for personalized suggestion generation.

Configuration

The server accepts the following environment variables:

API_KEY – Your Appwrite API key with appropriate project scopes.

MCP_PORT (optional) – Port on which the MCP server listens (default: 4000).

Available Tools/Resources

documents.list: Retrieve document lists from a specified collection.

storage.createFile: Upload files to Appwrite Storage buckets with permissions.

Who Should Use This?

This server is perfect for:

Use case 1: AI developers integrating LLMs with Appwrite backend services.

Use case 2: Teams building multi-agent workflows that require secure, standardized data access.

Use case 3: Enterprises needing a scalable, compliant solution for AI-driven applications.

Conclusion

With the Appwrite MCP Server, you get a standardized, secure, and extensible interface to Appwrite’s backend, streamlining AI and agentic development. Give it a try and simplify your integration process!

Check out the GitHub repository for more information and to contribute.