n-r-w/knowledgegraph-mcp
If you are the rightful owner of knowledgegraph-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
KnowledgeGraph MCP Server provides persistent memory for LLMs across conversations using a knowledge graph.
The KnowledgeGraph MCP Server is designed to enhance the capabilities of language learning models (LLMs) by providing them with a persistent memory system. This server allows LLMs like Claude or VS Code to remember and recall information about users, projects, and preferences through a structured knowledge graph. By utilizing multiple storage backends such as PostgreSQL and SQLite, the server ensures data persistence and project separation, allowing for isolated and organized data management. The server also features advanced search capabilities, including fuzzy search and pagination, to efficiently retrieve information. This setup is particularly beneficial for users who require a robust memory system to manage complex data interactions across different projects and platforms.
Features
- Multiple Storage Backends: Supports PostgreSQL and SQLite for flexible data storage.
- Project Separation: Automatically detects and isolates different projects.
- Advanced Search: Includes fuzzy search and pagination for efficient data retrieval.
- Easy Setup: Offers simple installation options via NPX or Docker.
- Persistent Memory: Provides LLMs with the ability to remember and recall information.
Tools
create_entities
Create new entities in the knowledge graph.
create_relations
Connect entities to enable discovery queries.
add_observations
Add factual observations to existing entities.
add_tags
Add status/category tags for instant filtering.
read_graph
Retrieve the complete knowledge graph.
search_knowledge
Search entities by text or tags.
open_nodes
Retrieve specific entities by exact names.
delete_entities
Permanently delete entities and their relationships.
delete_observations
Remove specific observations from entities.
delete_relations
Update relationship structure by removing connections.
remove_tags
Update entity status by removing outdated tags.