Memory MCP Server
@evangstav
Memory MCP Server について
概要はまだありません
基本情報
設定
ツール
ツールは検出されませんでした
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
概要
What is Memory MCP Server?
A Model Context Protocol (MCP) server that provides knowledge graph functionality for managing entities, relations, and observations in memory, with strict validation rules to maintain data consistency.
How to use Memory MCP Server?
Install the server in Claude Desktop with mcp install main.py -v MEMORY_FILE_PATH=/path/to/memory.jsonl. It offers tools to create, read, search, and delete entities and relations, and to flush the entire memory.
Key features of Memory MCP Server
- Strict validation rules for entity names, observations, and relations
- Eight supported entity types: person, concept, project, document, tool, organization, location, event
- Seven predefined relation types with no self-referential or circular dependencies allowed
- Natural language search with temporal queries and fuzzy matching
- Weighted search across entity names, types, and observations (80% similarity threshold)
- Typed responses with error types: NOT_FOUND, VALIDATION_ERROR, INTERNAL_ERROR, ALREADY_EXISTS, INVALID_RELATION
Use cases of Memory MCP Server
- Build and query a personal knowledge graph of people, projects, and documents
- Track dependencies between software tools and work initiatives
- Maintain a structured memory of events, locations, and organizations
- Search past observations using natural language (“most recent workout”)
FAQ from Memory MCP Server
What are the validation rules for entity names?
Entity names must start with a lowercase letter, contain only lowercase letters, numbers, and hyphens, be unique, and have a maximum length of 100 characters.
Where is the knowledge graph data stored?
Data is stored in a JSON Lines file at the path specified by the MEMORY_FILE_PATH environment variable.
What runtime dependencies does the server require?
The server is installed via mcp install into Claude Desktop and requires Python. Tests can be run with pytest tests/.
Are there any limits on observations or relations?
Observations must be non-empty, unique per entity, and at most 500 characters. Relations must be one of the predefined types; self-referential and circular relations are not allowed.
「メモリとナレッジ」の他のコンテンツ
MCP server for Obsidian
MarkusPfundsteinMCP server that interacts with Obsidian via the Obsidian rest API community plugin
Solomd
zhitongblogA markdown editor — and the bridge to your LLM. Local-first, MIT, ~15 MB. Bundled MCP server lets Claude Code / Codex / Cursor drive your vault directly. 14 AI providers BYOK.
Notion MCP Integration
danhilseA simple MCP integration that allows Claude to read and manage a personal Notion todo list
Mcp Knowledge Graph
shanehollomanMCP server enabling persistent memory for Claude through a local knowledge graph - fork focused on local development
JupyterMCP - Jupyter Notebook Model Context Protocol Integration
jjsantos01A Model Context Protocol (MCP) for Jupyter Notebook
コメント