Model Context Protocol (MCP) Integration

WeMush provides a fully-featured MCP server that enables AI assistants to access your mycology research data, specimens, and cultivation insights. Any MCP-compatible client can connect to WeMush.

What is MCP?

The Model Context Protocol (MCP) is an open standard developed by Anthropic that allows AI applications to securely connect to external data sources and tools. Think of it as a universal adapter between AI assistants and your data.

Why MCP Matters

• Open Standard: Not locked to any single AI vendor
• Secure: OAuth 2.0 with PKCE authentication
• Granular Permissions: You control exactly what data is shared
• Interoperable: Works with Claude, Cursor, Windsurf, Cline, and more

Compatible AI Clients

The WeMush MCP server works with any client that supports the Model Context Protocol:

Quick Setup

Claude Desktop / Claude.ai

The easiest way to connect - Claude has native MCP support:

1. Open Claude and navigate to Settings → Integrations
2. Find "WeMush" in the MCP integrations list
3. Click Connect and authorize with your WeMush account
4. Start chatting! Try: "What specimens do I have in colonization?"

Detailed Claude Setup Guide →

Cursor

Add WeMush to your Cursor MCP configuration:

{
  "mcpServers": {
    "wemush": {
      "url": "https://wemush.com/api/mcp/research/sse",
      "transport": "sse",
      "auth": {
        "type": "oauth",
        "authorizationUrl": "https://wemush.com/api/oauth/authorize",
        "tokenUrl": "https://wemush.com/api/oauth/token",
        "clientId": "wemush-mcp-public",
        "scopes": ["profile:read", "specimens:read", "research:read"]
      }
    }
  }
}

Windsurf

Add to your Windsurf MCP configuration (similar to Cursor):

{
  "mcpServers": {
    "wemush": {
      "url": "https://wemush.com/api/mcp/research/sse",
      "transport": "sse"
    }
  }
}

Then authenticate when prompted.

Cline / Continue

For VS Code extensions, add to your MCP settings:

{
  "mcp.servers": {
    "wemush": {
      "url": "https://wemush.com/api/mcp/research/sse"
    }
  }
}

Self-Hosted / Custom Clients

For custom integrations or self-hosted deployments:

Server URL: https://wemush.com/api/mcp/research

Transports Supported:

• SSE (Server-Sent Events): https://wemush.com/api/mcp/research/sse
• Streamable HTTP: https://wemush.com/api/mcp/research/mcp

OAuth Endpoints:

• Authorization: https://wemush.com/api/oauth/authorize
• Token: https://wemush.com/api/oauth/token
• Client ID: wemush-mcp-public

Note: We may introduce mcp.wemush.com as a shorter vanity URL in the future. The endpoints above are guaranteed to remain stable.

Available Capabilities

Once connected, the WeMush MCP server provides:

27 Tools

Organized into six categories for specimen management, research participation, and analytics:

See Full Tool Reference →

5 Prompts

Pre-built workflows for common tasks:

• Analyze Cultivation - Get AI analysis of your specimens
• Research Project Overview - Summarize any research project
• Strain Comparison - Compare cultivation requirements
• Troubleshoot Cultivation - Diagnose problems
• Summarize Activity - Review your contributions

3 Resources

Static resources available to connected clients:

• Cultivation Guide
• Research Glossary
• Your Research Profile

Authentication & Scopes

WeMush uses OAuth 2.0 with PKCE for secure authentication. During connection, you'll approve specific scopes:

You can revoke access at any time from WeMush Settings → Connected Apps.

Privacy & Security

K-Anonymity Protection

Research analytics use k=5 anonymity - data is only shown when at least 5 participants share similar characteristics.

Token Lifecycle

• Access tokens expire after 1 hour
• Refresh tokens expire after 30 days
• All access is logged for security auditing

Data Isolation

Each connection is scoped to your authorized organizations. AI clients cannot access data from other users or organizations.

Open Source

The WeMush MCP server is open source:

• Repository: github.com/wemush/wemush-skills
• License: MIT

You can self-host, customize, or contribute to the project.

Next Steps

• Detailed Tool Reference - Complete documentation of all 27 tools
• Open Standard (WOLS) - Learn about our data interoperability standard
• API Reference - REST API documentation (coming soon)

Support

• Email: support@wemush.com
• Status: wemush.com/status
• GitHub Issues: wemush/wemush-skills/issues