Claude Memory Architecture

Persistent memory management for Claude and other AI assistants. Give your AI a brain that remembers.

Born from the Claude Memory Architecture research project.

Quick Install

# Using uvx (recommended)
uvx mnemonic-memory

# Or pip
pip install mnemonic-memory

Claude Code Setup

Add to your Claude Code MCP config (~/.claude.json or project .claude/settings.json):

{
  "mcpServers": {
    "mnemonic": {
      "command": "uvx",
      "args": ["mnemonic-memory"]
    }
  }
}

Or with pip-installed version:

{
  "mcpServers": {
    "mnemonic": {
      "command": "mnemonic"
    }
  }
}

Features

Free Tier (Available Now)

Tool	Description
mnemonic_create	Create a new memory with tags and triggers
mnemonic_get	Retrieve a memory by ID
mnemonic_update	Update an existing memory
mnemonic_delete	Permanently delete a memory
mnemonic_search	Full-text search with tag filtering
mnemonic_list	List all memories with sorting
mnemonic_pin	Pin important memories
mnemonic_export	Export all memories to JSON
mnemonic_import	Import from a previous export
mnemonic_stats	View memory statistics

Premium Tier (Coming Soon) - $29 one-time or $5/mo

• Semantic Search: Find memories by meaning, not just keywords
• Auto-Decay: Memories fade based on access patterns
• Consolidation: Automatically merge similar memories
• Maintenance: Clean up stale data

Usage Examples

Create a memory

mnemonic_create({
  "title": "Bash's communication preferences",
  "content": "Direct and concise. Lead with conclusions (Minto Pyramid). No fluff.",
  "tags": ["preferences", "communication"],
  "triggers": ["how should I respond to bash"]
})

Search memories

# Full-text search
mnemonic_search({"query": "communication preferences"})

# Filter by tags
mnemonic_search({"tags": ["preferences"]})

# Combined
mnemonic_search({"query": "direct", "tags": ["communication"]})

Pin important memories

mnemonic_pin({"id": "abc-123", "pinned": true})

Data Storage

Memories are stored in ~/.mnemonic/memories.db (SQLite).

Override with MNEMONIC_DATA_DIR environment variable:

export MNEMONIC_DATA_DIR="/custom/path"

Schema

memories
├── id (TEXT, UUID)
├── type (TEXT, default 'memory')
├── title (TEXT)
├── content (TEXT)
├── weight (REAL, 0.1-1.0)
├── pinned (BOOLEAN)
├── emotional_flag (BOOLEAN)
├── created_at (TEXT, ISO timestamp)
├── updated_at (TEXT)
├── last_accessed_at (TEXT)
├── access_count (INTEGER)
├── status (TEXT: active, archived)
└── metadata (JSON)

tags
├── id (INTEGER)
└── name (TEXT, unique)

memory_tags (many-to-many)

triggers
├── id (INTEGER)
├── memory_id (TEXT)
└── phrase (TEXT)

Architecture

This project implements the Claude Memory Architecture research:

• Weighted memories: Not all memories are equal (0.1-1.0 weight)
• Memory decay: Unused things fade (premium feature)
• Depth on demand: Load summaries, expand when needed
• Trigger-based recall: Phrases that surface relevant memories

Development

# Clone
git clone https://github.com/bashoh/mnemonic-memory
cd mnemonic-memory

# Install with dev dependencies
pip install -e ".[dev]"

# Run locally
python -m mnemonic

License

MIT

Author

Built by Bash @ Wishly Group

---

"Current AI assistants have Alzheimer's. Let's fix that."