Memory Search MCP Server
Uses SQLite with FTS5 for data storage and fast full-text search, providing persistent memory storage in a local database file with zero external dependencies.
Click on "Install 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 Search MCP Serversearch for user's UI preferences from last week"
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.
By MEOK AI Labs — Sovereign AI tools for everyone.
Memory Search MCP Server
Persistent memory system for AI agents and assistants. Record episodic memories with care-weighting and emotional valence, search with full-text relevance ranking, maintain a knowledge base of facts and reference material, follow temporal chains, and consolidate old memories automatically.
Zero external dependencies beyond the mcp package -- uses SQLite with FTS5 for fast full-text search. Data persists in ~/.mcp-memory/memories.db.
Tools
Tool | Description |
| Store a memory episode with care weight, importance, emotion, and tags |
| Full-text semantic search with care weight and tag filtering |
| Add persistent facts/reference material to the knowledge base |
| Search the knowledge base by topic or content |
| Browse recent memories sorted by time, importance, or access count |
| Memory store statistics: counts, averages, storage size |
| Follow the timeline forward/backward from any memory |
| Archive old low-access memories to save space |
Installation
pip install mcpUsage
Run the server
python server.pyClaude Desktop config
{
"mcpServers": {
"memory-search": {
"command": "python",
"args": ["/path/to/memory-search-mcp/server.py"]
}
}
}Example calls
Record a memory:
Tool: record_memory
Input: {"content": "User prefers dark mode and compact layouts", "source_agent": "preferences", "memory_type": "insight", "care_weight": 0.8, "tags": ["preferences", "ui"]}
Output: {"success": true, "episode_id": "a3f2b1c8d9e0", "timestamp": "2026-04-13T10:30:00"}Search memories:
Tool: search_memory
Input: {"query": "user interface preferences", "limit": 5, "care_weight_min": 0.5}
Output: {"results": [...], "count": 3, "query": "user interface preferences"}Add knowledge:
Tool: add_knowledge
Input: {"topic": "Python asyncio", "content": "Use asyncio.gather() for concurrent coroutines...", "confidence": 0.9}
Output: {"success": true, "knowledge_id": "k1a2b3c4d5e6", "topic": "Python asyncio"}Follow temporal chain:
Tool: get_temporal_chain
Input: {"episode_id": "a3f2b1c8d9e0", "direction": "backward", "max_steps": 10}
Output: {"chain": [...], "direction": "backward", "steps": 7}Data Storage
All data is stored in ~/.mcp-memory/memories.db (SQLite). To reset, simply delete this file. To back up, copy it.
Pricing
Tier | Limit | Price |
Free | 100 calls/day | $0 |
Pro | Unlimited + vector embeddings + cloud sync | $9/mo |
Enterprise | Custom + team sharing + encryption at rest | Contact us |
License
MIT
This server cannot be installed
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/CSOAI-ORG/memory-search-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server