Memory Server MCP
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 Server MCP search for memories about the project launch timeline"
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 Server MCP
A MCP server for persistent memory storage with advanced features like tagging, timestamping, expiration, content search, and inter-memory linking. Built for integration with MCP clients like Claude for Desktop.
Features
Save/Update Memories (
save_memory): Store or update a memory with a unique key and value. Supports optional tags and expiration settings.key(string): Unique identifier for the memory.value(string): The content of the memory.tags(array of strings, optional): A list of tags to associate with the memory. If provided, replaces existing tags. Pass an empty array to remove all tags.expires_in_seconds(number, optional): The number of seconds until this memory expires. Pass0or a negative number to remove expiration.
Recall Memory (
recall_memory): Retrieve a memory by its key, including its value, associated tags, creation timestamp, last updated timestamp, and expiration timestamp.List Memories (
list_memories): List all non-expired memory keys. Optionally, filter memories by a specific tag.tag(string, optional): Filter memories by this tag.
Delete Memory (
delete_memory): Permanently remove a memory by its key. Also removes any links pointing to the deleted memory.List All Tags (
list_all_tags): Get a list of all unique tags currently in use across all non-expired memories.Search Memory Content (
search_memory_content): Search for memories whose content (value) contains a given query string (case-insensitive, non-expired memories only).Link Memories (
link_memories): Create a directed link from a source memory to a target memory with a specified relationship type.source_key(string): The key of the memory from which the link originates.target_key(string): The key of the memory to which the link points.relationship_type(string): The type of relationship (e.g.,related_to,depends_on,is_part_of).
Get Linked Memories (
get_linked_memories): Retrieve a list of memories linked from a given source memory, optionally filtered by relationship type.
Related MCP server: Simple Memory MCP
Installation & Usage
Quick Start with npx (Recommended)
npx memory-server-mcpGlobal Installation
npm install -g memory-server-mcp
memory-server-mcpUsage with MCP Clients
Claude for Desktop
Add MCP Server Configuration: In your Claude for Desktop settings, add a new MCP server:
{ "mcpServers": { "memory": { "command": "npx", "args": ["memory-server-mcp"] } } }Restart Claude for Desktop to load the new server.
Start using memory commands in your conversations with Claude.
Example Usage
Once connected to an MCP client, you can use commands like:
Save a memory about my project:
Key: project_status
Value: Working on memory server MCP, almost ready for release
Tags: work, mcp, nodejs
Recall what I saved about my project status.
Search for memories containing "nodejs".
Link my project memory to related technologies.Available Tools
12 toolsdelete_memoryC
Delete a memory by key
Inputs: key (required).
find_memories_within_degreesC
Find all memories within N degrees of separation from a starting memory
Inputs: start_key (required), max_degrees, relationship_type.
find_memory_pathB
Find the shortest path between two memories through their links
Inputs: from_key (required), to_key (required).
get_linked_memoriesC
Get all memories linked from a source memory
Inputs: source_key (required), relationship_type.
link_memoriesC
Create a link between two memories with a relationship type
Inputs: source_key (required), target_key (required), relationship_type (required).
list_all_tagsA
List all unique tags used across all memories
Inputs: no named parameters.
list_memoriesB
List all memory keys, optionally filtered by tag
Inputs: tag.
recall_memoryB
Retrieve a value by key from memory, with optional related memories
Inputs: key (required), include_related, max_related.
save_memoryC
Save a value with a key to memory, optionally with tags and expiration
Inputs: key (required), value (required), tags, expires_in_seconds.
search_by_relationshipC
Search for memories based on relationship patterns
Inputs: pattern (required).
search_memory_contentC
Search memory content using fuzzy search
Inputs: query (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 12 tools
Most tools have distinct purposes, but there is some overlap between 'find_related_memories' and 'search_by_relationship', as both involve relationship-based queries, which could cause confusion. Other tools like 'recall_memory' and 'get_linked_memories' are clearly differentiated, with the former retrieving a single memory and the latter focusing on linked connections.
All tool names follow a consistent snake_case pattern with clear verb_noun structures, such as 'delete_memory', 'save_memory', and 'search_memory_content'. This uniformity makes the tool set predictable and easy to navigate, with no deviations in naming conventions.
With 12 tools, the count is well-suited for a memory management server, covering core operations like save, recall, delete, search, and linking. Each tool serves a specific function without redundancy, making the set comprehensive yet manageable for its domain.
The tool set provides complete coverage for memory management, including CRUD operations (save, recall, delete, list), advanced features like linking, searching by content and relationships, and utility functions such as tag listing. There are no obvious gaps, supporting a full lifecycle from creation to complex queries.
Maintenance
Related MCP Connectors
Persistent memory for AI agents. Search and store durable facts, preferences and decisions.
Persistent memory for AI assistants: store, search, and connect knowledge across conversations.
Persistent cloud memory for AI agents. Store and search key-value memories across sessions.
Persistent memory for AI agents: add, search, update, and delete long-term memories.
Related MCP Servers
- FlicenseAqualityNot gradedmaintenanceProvides long-term memory storage for AI assistants with semantic search, enabling persistent storage of preferences, decisions, and context with relationship tracking between memories.19-
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to store and retrieve persistent memories with a web management interface. Supports creating, searching, and managing memories through natural language commands or a visual web dashboard.27 npm1MIT
- FlicenseBqualityDmaintenanceEnables persistent memory for AI systems by providing tools for episodic, semantic, and procedural data storage through a vector-and-graph-enhanced database. It allows models to maintain long-term continuity using similarity search, thematic clustering, and identity tracking.241-

Oceanir Memoryofficial
FlicenseNot gradedqualityDmaintenanceProvides persistent long-term memory for AI coding agents by storing entities, relations, and observations across different sessions. It enables users to manage and query structured knowledge like coding preferences, project patterns, and technical solutions via a graph-based storage system.1-