Memory MCP Server
Supports installation on macOS systems, with specific configuration paths for Claude for Desktop integration.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Memory MCP Server store that I need to call Sarah about the project tomorrow with tags work, followup"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Memory MCP Server
A Model Context Protocol (MCP) server that provides tools for storing, retrieving, and managing memories saved to a local file.
Features
Store text memories with optional tags
Search memories by text or tags
Retrieve all stored memories
Delete specific memories
View tag usage statistics
Persistent storage in a local JSON file
Related MCP server: Claude Continuity MCP Server
Installation
From Source
Clone this repository
Install dependencies:
npm installBuild the server:
npm run build
Usage
Running the Server
Run the server with:
npm start -- [storage-path]Where storage-path is the directory where you want to store the memories file (defaults to the current directory).
Setting up with Claude for Desktop
To use this server with Claude for Desktop:
Open your Claude for Desktop configuration file:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.json
Add the server configuration:
{
"mcpServers": {
"memory": {
"command": "node",
"args": [
"/path/to/mcp-memory-server/build/index.js",
"/path/to/storage/directory"
]
}
}
}Restart Claude for Desktop. You may need to force-quit any running Claude processes for the changes to take effect.
Available Tools
The server exposes the following tools:
store-memory
Stores a new memory.
Parameters:
content(string, required): The text content to storetags(array of strings, optional): Tags to categorize the memory
search-memories
Searches for memories.
Parameters:
searchTerm(string, optional): Text to search for in the memory contenttag(string, optional): Filter memories by this tag
get-all-memories
Retrieves all stored memories.
Parameters: None
delete-memory
Deletes a memory by ID.
Parameters:
id(string, required): ID of the memory to delete
get-tags-summary
Gets a summary of all used tags.
Parameters: None
Example Usage in Claude
Here are some example queries to use with Claude:
"Store this information: The project deadline is March 15, 2025. Tags: work, deadlines"
"What memories do I have related to work?"
"Show me all my memories"
"Delete memory with ID abc123"
"What tags have I used in my memories?"
File Structure
The server stores memories in a file named memories.json in the specified storage directory.
License
MIT
Available Tools
14 toolsadd_observationsB
Add new observations to existing entities in the knowledge graph
Inputs: observations (required).
create_entitiesB
Create multiple new entities in the knowledge graph
Inputs: entities (required).
create_relationsB
Create multiple new relations between entities in the knowledge graph. Relations should be in active voice
Inputs: relations (required).
delete_entitiesA
Delete multiple entities and their associated relations from the knowledge graph
Inputs: entityNames (required).
delete_observationsB
Delete specific observations from entities in the knowledge graph
Inputs: deletions (required).
delete_relationsA
Delete multiple relations from the knowledge graph
Inputs: relations (required).
get_frequently_usedC
Find entities that are accessed frequently
Inputs: minAccessCount (required), entityType.
get_stale_entitiesC
Find entities that haven't been updated recently
Inputs: days (required), entityType.
open_nodesC
Open specific nodes in the knowledge graph by their names
Inputs: names (required).
read_graphB
Read the entire knowledge graph
Inputs: no named parameters.
search_nodesC
Search for nodes in the knowledge graph based on a query
Inputs: query (required).
update_entityC
Update an existing entity, preserving previous observations
Inputs: name (required), metadata, observations.
validate_component_propsC
Validate component props to prevent hallucination. Checks if props exist in @props observations
Inputs: componentName (required), propsToCheck (required).
verify_graph_integrityA
Verify the integrity of the knowledge graph by checking for orphaned entities in relationships. Returns hallucinated entity names with fuzzy search suggestions.
Inputs: maxSuggestions.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 14 tools
Most tools have distinct purposes targeting specific operations on entities, relations, or observations, with clear boundaries. However, 'open_nodes' and 'search_nodes' could be confused as both involve accessing nodes, though 'open_nodes' appears to retrieve by name while 'search_nodes' uses a query.
Tool names follow a consistent verb_noun pattern throughout, such as 'add_observations', 'create_entities', and 'delete_relations'. All tools use snake_case with clear, descriptive verbs, making the set predictable and readable.
With 14 tools, the count is well-scoped for a knowledge graph management server, covering CRUD operations, querying, and maintenance tasks. Each tool earns its place without feeling excessive or insufficient for the domain.
The toolset provides comprehensive coverage for knowledge graph operations, including entity and relation management, observation handling, search, and integrity checks. A minor gap is the lack of a tool to update relations, but agents can work around this by deleting and recreating them.
Maintenance
Related MCP Connectors
Persistent memory layer for AI tools. Save and recall notes across Claude and other MCP clients.
Persistent memory for AI assistants. Save once; recall from Claude, ChatGPT, or any MCP client.
Persistent memory for AI agents across Claude, ChatGPT and any MCP client.
Private persistent memory for Claude, ChatGPT & Gemini via MCP - semantic search, zero-code setup.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceA Model Context Protocol server for Claude Desktop that provides structured memory management across chat sessions, allowing Claude to maintain context and build a knowledge base within project directories.7MIT
- AlicenseNot gradedqualityDmaintenanceA Model Context Protocol server that provides persistent memory and conversation continuity for Claude Desktop and Claude Code, allowing users to save and restore project context when threads hit token limits.MIT
- AlicenseAqualityDmaintenancePersistent memory MCP server that allows Claude to store, organize, and retrieve knowledge across sessions without consuming context window tokens.2482 PyPI18MIT
- FlicenseNot gradedqualityDmaintenanceA Model Context Protocol (MCP) server that gives Claude a persistent personal notebook. Notes are stored on disk as JSON, so they survive restarts and are shared across every tool and resource.-