mcp
Agentset MCP Server - Build RAG with Agentic superpowers
claude mcp add --transport stdio agentset-ai-mcp-server npx -y @agentset/mcp@latest \ --env AGENTSET_API_KEY="your_api_key" \ --env AGENTSET_NAMESPACE_ID="your_namespace_id"
How to use
This MCP server integrates Agentset's retrieval-augmented generation tooling, enabling you to expose a configurable set of tools that can be used by an agent for document-based tasks. By adding this server to your Claude environment, you enable the agent to access Agentset's capabilities (such as querying documents, performing RAG-style retrieval, and leveraging the tool descriptions you provide) via the MCP interface. The included example shows how to add the Agentset MCP to Claude so that the agent can call the @agentset/mcp package as a managed tool within your namespace.
To use, ensure you have an API key and a namespace ID for your Agentset account. Then run the MCP via npx, which will fetch the latest package and launch the server under the specified namespace. You can customize tool descriptions and behavior by passing descriptive text or additional flags when invoking the MCP, and you can set environment variables to control authentication and scope for each request.
How to install
Prerequisites:
- Node.js and npm (or use a compatible environment that can run npx)
- Access credentials for Agentset (API key and namespace ID)
Installation steps:
-
Ensure you have an API key and namespace ready for Agentset:
- AGENTSET_API_KEY=your-api-key
- AGENTSET_NAMESPACE_ID=your-namespace-id
-
Install and run the MCP via npx (as shown in the repository):
# Replace with your actual API key and namespace if you want to export them inline
AGENTSET_API_KEY=your-api-key AGENTSET_NAMESPACE_ID=your-namespace-id npx @agentset/mcp --ns your-namespace-id
- If you are integrating with Claude, use the provided Claude config snippet to add the MCP as a tool:
{
"mcpServers": {
"agentset": {
"command": "npx",
"args": ["-y", "@agentset/mcp@latest"],
"env": {
"AGENTSET_API_KEY": "agentset_xxx",
"AGENTSET_NAMESPACE_ID": "ns_xxx"
}
}
}
}
- Verify the server is reachable and the tool is discoverable by your agent platform. Ensure firewall and network rules allow MCP endpoint access if running in a restricted environment.
Additional notes
Tips and common notes:
- Environment variables: AGENTSET_API_KEY and AGENTSET_NAMESPACE_ID are required to access Agentset resources. You can override them per invocation as needed.
- Tool descriptions: You can customize the tool description passed to the MCP to help users understand its purpose within your app (as shown by the -d option in examples).
- Tenant handling: You can specify a tenant via -t your-tenant-id when invoking the MCP for multi-tenant setups.
- If you encounter issues, check that the API key has the correct permissions for the target namespace and that the npm package @agentset/mcp is accessible from your environment.
- Keeping the MCP up to date is as simple as using the latest tag in the npx command (@latest).
Related MCP Servers
zen
Selfhosted notes app. Single golang binary, notes stored as markdown within SQLite, full-text search, very low resource usage
MCP -Deepseek_R1
A Model Context Protocol (MCP) server implementation connecting Claude Desktop with DeepSeek's language models (R1/V3)
mcp-fhir
A Model Context Protocol implementation for FHIR
mcp
Inkdrop Model Context Protocol Server
mcp-appium-gestures
This is a Model Context Protocol (MCP) server providing resources and tools for Appium mobile gestures using Actions API..
dubco -npm
The (Unofficial) dubco-mcp-server enables AI assistants to manage Dub.co short links via the Model Context Protocol. It provides three MCP tools: create_link for generating new short URLs, update_link for modifying existing links, and delete_link for removing short links.