Self-hosted MCP＃
Deploy kluster.ai's MCP server locally for development and testing. This self-hosted implementation gives you full control over your infrastructure while providing the same verification tools as Cloud MCP.
Prerequisites＃
Before deploying the self-hosted MCP server, ensure you have:
- A kluster.ai account: Sign up on the kluster.ai platform if you don't have one.
- A kluster.ai API key: After signing in, go to the API Keys section and create a new key. For detailed instructions, check out the Get an API key guide.
- A runtime environment: You can use either Docker Desktop or Node.js 18+.
- Git: For cloning the repository.
Clone repository＃
First, clone the MCP server repository:
git clone https://github.com/kluster-ai/verify-mcp
cd verify-mcp
Deployment options＃
Build and run with Docker:
docker build -t kluster-verify-mcp .
docker run --rm -p 3001:3001 kluster-verify-mcp --api-key YOUR_API_KEY
Install dependencies and run:
npm install
npm run build
npm start -- --api-key YOUR_API_KEY
The server will start on http://localhost:3001 with the MCP endpoint at /stream.
Client integration＃
Once your self-hosted server is running, configure your AI clients using the Client integrations guide.
Use these connection details:
- MCP endpoint:
http://localhost:3001/stream. - Authentication: Your kluster.ai API key.
Available tools＃
Your self-hosted deployment provides the same verification tools as Cloud MCP:
verify: Validates claims against reliable online sources.verify_document: Verifies claims about uploaded document content.
For detailed parameters and response formats, see the Tools reference.
Next steps＃
- Configure clients: Follow the Client integrations guide for VS Code, Claude Desktop, and other platforms.
- Learn the tools: See Tools reference for detailed examples.
- Try Cloud MCP: Consider Cloud MCP for managed cloud deployment.