Memory-Man
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-Manstore that we use JWT with Redis for auth"
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-Man 🧠
A persistent memory system for Claude Code that solves the context window problem by storing and retrieving project knowledge across sessions.
What is Memory-Man?
Memory-Man is an MCP (Model Context Protocol) server that acts as Claude's long-term memory. Instead of losing context when the conversation window fills up, Claude can now:
Store important information about your projects
Search through past conversations and decisions
Retrieve relevant context when working on similar problems
Track patterns across multiple projects
Related MCP server: Claude Memory MCP Server
Features
🗂️ Project-Aware Storage - Memories are organized by project
🔍 Smart Search - Find relevant memories by content, tags, or category
📊 Usage Analytics - Track which memories are most valuable
🏷️ Intelligent Tagging - Categorize memories for easy retrieval
⚡ Fast SQLite Backend - Efficient local storage
🔧 Easy Integration - Works with Claude Code out of the box
Quick Start
Install Dependencies:
source venv/bin/activate pip install -r requirements-core.txtTest the Server:
python test_server.pyConfigure Claude Code (see Integration Guide below)
Memory Categories
architecture - System design decisions and patterns
setup - Project configuration and environment
bug_fix - Solutions to bugs and issues
todo - Future work and planned features
pattern - Reusable code patterns and best practices
command - Useful commands and scripts
Integration with Claude Code
Method 1: Global MCP Configuration
Add to your global Claude Code config (~/.claude/mcp_servers.json):
{
"mcpServers": {
"memory-man": {
"command": "python",
"args": ["-m", "memory_man"],
"env": {
"PYTHONPATH": "/home/beano/DevProjects/python/memory-man/src"
}
}
}
}Method 2: Project-Specific Configuration
Add mcp_config.json to any project:
{
"mcpServers": {
"memory-man": {
"command": "python",
"args": ["-m", "memory_man"],
"env": {
"PYTHONPATH": "/home/beano/DevProjects/python/memory-man/src"
}
}
}
}Usage Examples
🚀 Smart Auto-Store (Recommended)
# Just provide the content - everything else is auto-detected!
memory_auto_store(
content="We use JWT tokens with Redis for session management. Tokens expire after 24 hours."
)
# Auto-detects:
# - Project: current directory name
# - Category: architecture (based on content)
# - Tags: ["jwt", "redis", "auth", "database"]
# - Framework: detected from project files🔍 Smart Project Detection
# Analyze your current project
project_detect()
# Returns project info + suggestions:
# - Technology stack (Python, JavaScript, etc.)
# - Framework (Django, React, FastAPI, etc.)
# - Recommended memory categories
# - Setup suggestions🧠 Context-Aware Suggestions
# Get relevant memories for current work
memory_suggest_related(context="authentication setup")
# Returns:
# - Related memories from current project
# - Similar patterns from other projects
# - Smart suggestions based on context📝 Manual Storage (When You Need Control)
# Store architecture decision manually
memory_store(
content="We use Redis for session storage with 24-hour TTL",
category="architecture",
project="my-web-app",
tags=["redis", "sessions", "auth"],
importance=8
)
# Search memories
memory_search(query="authentication", project="my-web-app")
# Get project overview
project_summary(project="my-web-app")Available Tools
Core Tools
Tool | Description |
| Store new memories with project context |
| Search memories by query, project, or category |
| Get a specific memory by ID |
| Update existing memories |
| Delete memories |
| Get summary of a project's memories |
| List all projects with memories |
Smart Tools ✨
Tool | Description |
| Auto-detect project, category, and tags |
| Analyze current project and get suggestions |
| Find relevant memories for current context |
Advanced Tools 🔧
Tool | Description |
| Generate intelligent project summaries |
| Analyze and optimize memory usage |
| Find memories ready for archival |
| Archive memories manually |
| Restore archived memories |
| Automatic memory lifecycle management |
Configuration
Environment variables (optional):
# Database location
MEMORY_MAN_DATABASE_URL=sqlite+aiosqlite:///./data/memories.db
# Storage directory
MEMORY_MAN_DATA_DIR=./data
# Search settings
MEMORY_MAN_SEARCH_LIMIT=20
MEMORY_MAN_MAX_MEMORY_SIZE=10000
# Logging
MEMORY_MAN_LOG_LEVEL=INFO
MEMORY_MAN_DEBUG=falseDevelopment
Running Tests
python test_server.pyCode Quality
# Format code
black src/
# Lint code
ruff check src/
# Type checking
mypy src/Adding New Features
Models: Add to
src/memory_man/models/Tools: Add to
src/memory_man/server.pyConfig: Update
src/memory_man/config.py
Troubleshooting
Server Won't Start
Check Python path in MCP configuration
Ensure virtual environment is properly set up
Verify all dependencies are installed
Memory Search Issues
Check if database exists in
data/memories.dbVerify project names match exactly
Try broader search terms
Performance Issues
Database will be created automatically on first run
Consider adding indexes for large datasets
Use importance levels to prioritize memories
Contributing
This is designed as a personal tool for your Claude Code workflow. Feel free to modify:
Add new memory categories
Implement semantic search
Add memory expiration
Create backup/sync features
License
MIT License - Feel free to use and modify as needed.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- Alicense-qualityAmaintenanceMCP server providing persistent, searchable memory management for AI workflows, enabling Claude Code to store, retrieve, and organize context through CRUD operations and knowledge tools.MIT
- -license-qualityBmaintenanceEnterprise-grade MCP server for persistent, intelligent memory management across Claude Code sessions.
- Alicense-qualityDmaintenanceAn MCP server that gives Claude persistent memory by storing conversation context, entities, and enabling semantic search across sessions.201MIT
- Alicense-qualityDmaintenancePersistent memory MCP server for Claude Code that captures and recalls project context across sessions, eliminating the need to re-explain architecture and decisions daily.2441MIT
Related MCP Connectors
Cloud-hosted MCP server for durable AI memory
User-owned memory for AI agents, Copilot, Claude, IDEs, CLIs, and chat apps over remote MCP.
Persistent memory and cross-session learning for AI coding assistants (hosted remote MCP).
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
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/HelyeFab/memory-man'
If you have feedback or need assistance with the MCP directory API, please join our Discord server