Agentset MCP

What is Agentset MCP?

Agentset MCP is a Model Context Protocol server for Agentset, an open-source platform for Retrieval-Augmented Generation (RAG). It is designed for developers who want to build intelligent, document-based applications quickly and efficiently.

How to use Agentset MCP?

Install and run using npm, yarn, or pnpm with an AGENTSET_API_KEY environment variable and a namespace ID passed with --ns. To add to Claude, configure the server with the npx command and environment variables AGENTSET_API_KEY and AGENTSET_NAMESPACE_ID. Additional options include custom tool descriptions (-d) and tenant IDs (-t).

Key features of Agentset MCP

• MCP server for Agentset RAG platform
• Configurable namespace ID for data isolation
• Custom tool description support via -d flag
• Tenant ID support via -t flag
• Easy installation with npm, yarn, or pnpm
• Direct integration with Claude MCP configuration

Use cases of Agentset MCP

• Integrate RAG capabilities into Claude or other MCP clients
• Build document-based question-answering applications
• Isolate data access by namespace or tenant
• Customize tool descriptions for specific document workflows

FAQ from Agentset MCP

How do I install Agentset MCP?

Run AGENTSET_API_KEY=your-api-key npx @agentset/mcp --ns your-namespace-id (or use yarn/pnpm equivalents).

What environment variables are required?

AGENTSET_API_KEY (your API key) and AGENTSET_NAMESPACE_ID (your namespace ID) are required.

How do I add Agentset MCP to Claude?

Add the following to your MCP server configuration:

{
  "mcpServers": {
    "agentset": {
      "command": "npx",
      "args": ["-y", "@agentset/mcp@latest"],
      "env": {
        "AGENTSET_API_KEY": "agentset_xxx",
        "AGENTSET_NAMESPACE_ID": "ns_xxx"
      }
    }
  }
}

Can I specify a custom tool description?

Yes, use the -d flag followed by your description, e.g., -d "Your custom tool description".

Can I specify a tenant ID?

Yes, use the -t flag followed by your tenant ID, e.g., -t your-tenant-id.