MCP Server

NVIDIA AI CKG MCP Server

Name: NVIDIA AI CKG MCP Server
Author: io.github.Yarmoluk

io.github.Yarmoluk/ckg-nvidia-ai

What is the NVIDIA AI CKG MCP server?

NVIDIA AI knowledge graphs — 20 domains. 4x F1, 11x fewer tokens, SHA-256 provenance. MCP-native.

Copy-paste configuration for popular MCP clients.

transport: stdio

Config generated by PluginBench — verify against the source before use.

Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "ckg-nvidia-ai": {
      "command": "uvx",
      "args": [
        "ckg-nvidia-ai"
      ]
    }
  }
}

Cursor

~/.cursor/mcp.json

{
  "mcpServers": {
    "ckg-nvidia-ai": {
      "command": "uvx",
      "args": [
        "ckg-nvidia-ai"
      ]
    }
  }
}

Windsurf

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "ckg-nvidia-ai": {
      "command": "uvx",
      "args": [
        "ckg-nvidia-ai"
      ]
    }
  }
}

VS Code

.vscode/mcp.json

{
  "servers": {
    "ckg-nvidia-ai": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "ckg-nvidia-ai"
      ]
    }
  }
}

Claude Code

claude mcp add ckg-nvidia-ai -- uvx ckg-nvidia-ai

Active

97-domain knowledge graphs for AI agents — 4× F1 of RAG, 11× fewer tokens. MCP-native.

Python

MIT

Salesforce AgentForce knowledge graph — 40 nodes, Einstein Trust Layer, 11x fewer tokens than RAG.

NVIDIA Nemotron + Perplexity Sonar as knowledge graphs — 2 domains, 83 nodes. MCP-native.

NVIDIA NemoClaw knowledge graph — 55 nodes, F1 0.576 (+269% vs RAG), 11x fewer tokens. MCP-native.