What is a Knowledge Graph MCP Server?

A Knowledge Graph MCP (Model Context Protocol) Server is a service for creating, managing, analyzing, and visualizing knowledge graphs. It complies with the MCP standard, enabling seamless integration with AI assistants like Claude.

What are the core features of this MCP Server?

The server supports multiple graph types (topology, timelines, etc.), complete error handling, resource management for SVG and Markdown, and version status management.

How do I install the Knowledge Graph MCP Server?

You can install it via Smithery or manually using npm or pnpm. The installation guide provides detailed instructions for both methods.

What are the requirements for installing the MCP Server?

You need Node.js version 16.0.0 or higher and pnpm version 7.0.0 or higher.

How do I configure the server to work with Claude Desktop?

Add the server configuration to your claude_desktop_config.json file, specifying the command, arguments, and environment variables, including the path to your knowledge graph directory.

How do I use the MCP Server in Cursor?

Add the server configuration to your Cursor config file, similar to Claude Desktop, specifying the command, arguments, and environment variables.

What is Smithery?

Smithery is a tool for automatically installing the knowledge-graph-mcp for Claude Desktop.

Where are the prompt files located, and how do I use them?

Prompt files are in the /src/rules/prompts directory. Copy them to your Cursor rules directory and add them as default rules for Agent mode.

How can I trigger knowledge graph queries in Cursor’s Agent mode?

Type /ck in the editor to invoke the prompt defined in @graph-query.mdc. The Agent will analyze the context, query the knowledge graph, and integrate results into the conversation.

What graph management tools are available?

The server provides tools like create_graph, list_graphs, and publish_graph for creating, listing, and publishing knowledge graphs.

What node management tools are included?

Node management tools include add_node, update_node, delete_node, and get_node_details for managing nodes within a graph.

What edge management tools are available?

Edge management tools include add_edge, update_edge, and delete_edge for managing relationships between nodes.

What resource management tools are provided?

Resource management tools include get_creation_guidelines, save_resource, update_resource, delete_resource, and unlink_resource for managing resources like SVG and Markdown files.

How does the server handle errors?

All errors are logged to the md/error_log.txt file with timestamps, error messages, and stack traces.

What is MCP?

MCP (Model Context Protocol) is an open protocol that standardizes how applications provide context to LLMs.

How UBOS related to MCP Server?

UBOS is Full-stack AI Agent Development Platform. UBOS focused on bringing AI Agent to every business department. Our platform help you orchestrate AI Agents, connect them with your enterprise data, build custom AI Agents with your LLM model and Multi-Agent Systems.