$ cat atproto-mcp.md

4 min read
mcpblueskyatprotocolaillmtypescriptsocial-networking

Overview

The AT Protocol MCP Server is a comprehensive implementation of the Model Context Protocol (MCP) that provides LLMs with direct access to the AT Protocol ecosystem. This enables AI assistants like Claude to seamlessly interact with Bluesky and other AT Protocol-based social networks through natural language.

Key Innovation

Unlike traditional API wrappers, this MCP server is specifically designed for LLM consumption. Users don’t interact with the server directly—instead, they communicate with their LLM client in natural language (e.g., “search for posts about AI”), and the LLM client uses this MCP server to fulfill those requests via the MCP protocol.

Architecture

The server acts as a bridge between LLM clients and the AT Protocol ecosystem:

User → LLM Client (Claude Desktop) → MCP Protocol → This Server → AT Protocol → Bluesky

This architecture enables:

  • Natural Language Interface: Users interact in plain English
  • Intelligent Tool Selection: LLMs choose the appropriate AT Protocol operations
  • Seamless Integration: Works with any MCP-compatible AI assistant

Features

🔓 Dual Mode Operation

  • Unauthenticated Mode: Access public data without setup

    • View user profiles
    • Manage OAuth authentication flows

  • Authenticated Mode: Full functionality with app passwords or OAuth

    • Create, edit, and delete posts
    • Follow/unfollow users
    • Like and repost content
    • Access personalized timelines and notifications
    • Manage lists and moderation settings

🛠️ Comprehensive Tool Suite

30+ MCP Tools across multiple categories:

  • Social Operations: Create posts, reply, like, repost, follow/unfollow
  • Data Retrieval: Search posts, get followers/following, view threads and feeds
  • Content Management: Upload media, delete posts, update profiles, generate link previews
  • List Management: Create and manage user lists
  • Moderation: Mute, block, and report users/content
  • Real-time Streaming: Live event streams and notifications

🔐 Flexible Authentication

  • App Passwords: Simple setup for development
  • OAuth: Production-ready authentication flow
  • Optional: Works perfectly without authentication for public data access

Technical Implementation

Built With Modern Standards

  • TypeScript: Strict type checking for reliability
  • @atproto/api: Official AT Protocol SDK
  • @modelcontextprotocol/sdk: MCP specification compliance
  • Production Ready: Docker support, Kubernetes manifests, comprehensive monitoring

Security & Performance

  • Rate Limiting: Built-in respect for AT Protocol rate limits
  • Input Validation: Comprehensive sanitization and security safeguards
  • Connection Pooling: Optimized performance with caching
  • Health Monitoring: Prometheus metrics and Grafana dashboards

Use Cases

For LLM Clients

  • Claude Desktop: Direct integration for Bluesky interaction
  • Custom AI Applications: Build LLM-powered social media tools
  • Research Tools: Analyze AT Protocol networks with AI assistance

For Developers

  • Extend Functionality: Modular architecture for custom tools
  • Production Deployment: Enterprise-ready with Docker and Kubernetes
  • Open Source: Contribute to the growing MCP ecosystem

Getting Started

Quick Installation

# Install globally
npm install -g atproto-mcp

# Or use with npx
npx atproto-mcp

Claude Desktop Integration

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "atproto": {
      "command": "npx",
      "args": ["atproto-mcp"]
    }
  }
}

For authenticated mode with full functionality:

{
  "mcpServers": {
    "atproto": {
      "command": "npx",
      "args": ["atproto-mcp"],
      "env": {
        "ATPROTO_IDENTIFIER": "your-handle.bsky.social",
        "ATPROTO_PASSWORD": "your-app-password"
      }
    }
  }
}

Example Interactions

Once configured, you can ask Claude:

  • “Search for posts about artificial intelligence on Bluesky”
  • “Show me the profile of @example.bsky.social”
  • “Create a post about my latest project”
  • “Who are the followers of @popular-account.bsky.social?”
  • “Get my personalized timeline”

Production Deployment

The server includes comprehensive production features:

  • Docker Support: Multi-stage builds with security best practices
  • Kubernetes Ready: Helm charts and deployment manifests
  • Monitoring: Health checks, metrics, and logging
  • Security: Input sanitization, rate limiting, credential management

Development

Testing

# Run all tests
pnpm test

# Run with coverage
pnpm test:coverage

# Integration tests (connects to real AT Protocol servers)
pnpm test:integration

Quality Assurance

  • Full Type Checking: TypeScript with strict mode
  • Comprehensive Testing: Unit and integration tests
  • CI/CD Pipeline: Automated testing and deployment
  • Code Quality: ESLint and Prettier integration

Community & Support

Impact

This project bridges the gap between AI assistants and decentralized social networks, enabling:

  • Accessible Social Media: Natural language interaction with Bluesky
  • Developer Productivity: Build AT Protocol tools faster with AI assistance
  • Protocol Adoption: Lower barriers to AT Protocol integration
  • Open Ecosystem: Contribute to the growing MCP server ecosystem

Future Roadmap

  • Enhanced streaming capabilities
  • Advanced content filtering
  • Multi-account support
  • Custom feed algorithms
  • Extended moderation tools

Status: Production Ready
License: MIT
Maintained: Active Development