Stores AI memories as Obsidian-compatible Markdown files, enabling knowledge graph visualization.
Sponsored
This tool functions as a Model Context Protocol (MCP) server, designed to manage an AI's long-term memory. It adapts Anthropic's original memory server to store AI-generated insights, entities, and relationships directly as individual Markdown files. By leveraging Obsidian's `[[link]]` syntax and YAML frontmatter, it facilitates the visualization of these memories as an interactive knowledge graph, making the AI's learned information easily navigable and searchable within an Obsidian vault.
Key Features
011 GitHub stars
02Builds a knowledge graph of entities, relations, and observations
03Utilizes YAML frontmatter for metadata storage
04Provides search functionality across all stored memories
05Stores each entity as an individual Markdown file
06Integrates with Obsidian using `[[link]]` syntax for graph visualization
Use Cases
01Visualizing an AI's learned knowledge as an interactive graph in Obsidian
02Managing and organizing an AI's long-term memory
03Enabling an AI to access and query its past interactions and facts
