Knowledge Graph

Provides a Python-based MCP server to build, query, and visualize a knowledge graph with spaced repetition and mastery tracking.

python

0

GitHub Stars

python

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "zcsabbagh-knowledge-graph-mcp": {
      "command": "python",
      "args": [
        "-m",
        "knowledge_graph_mcp.server"
      ]
    }
  }
}

You can run and use the Knowledge Graph MCP Server to build, query, and manage a personalized knowledge map with spaced repetition. This server stores concepts and relationships in a graph, tracks mastery across recall, application, and explanation, and helps you identify knowledge gaps and misconceptions for targeted remediation.

How to use

You interact with the Knowledge Graph MCP Server through a client that supports the MCP protocol. Start the server locally, then connect your MCP client using the provided command. Once connected, you can create concept nodes, establish relationships, record learning progress, query for ready-to-learn concepts, and visualize learning paths or subgraphs.

How to install

Prerequisites: Python 3.10 or later, and a working Python environment.

Option 1: Install from source and run locally

git clone https://github.com/zcsabbagh/knowledge-graph-mcp.git
cd knowledge-graph-mcp
pip install -e .

# Start the server
python -m knowledge_graph_mcp.server

Option 2: Install from Smithery (if you prefer a hosted or client-only workflow)

Use the Smithery CLI to install the MCP package for your client setup.

Note: If you use an integrated Claude workflow, you can configure the client to point to the running server using the standard MCP connection settings described by your client.

Additional configuration and usage notes

The server starts with a straightforward Python module invocation. There is no additional configuration required for basic operation beyond starting the server. You can later extend functionality by adding nodes, edges, and queries through your MCP client.

To run from the project root, execute: python -m knowledge_graph_mcp.server

Available tools

add_node

Create a new concept node with properties such as concept name, description, domain, difficulty, and tags.

add_edge

Create a relationship between two concepts with a specified relation type (prerequisite, builds_on, related_to, etc.).

update_node

Update mastery scores and record reviews. Accepts a quality rating to trigger spaced repetition scheduling and optional misconception notes.

query_graph

Execute intelligent queries to uncover prerequisites, ready-to-learn concepts, due reviews, struggles, misconceptions, and knowledge gaps.

read_subgraph

Get the neighborhood around a concept and optionally generate Mermaid visualizations.

get_learning_path

Retrieve ordered prerequisites required to reach a target concept.

get_statistics

Obtain learning progress metrics for a domain.