mem0 Plugin for Cursor

Persistent memory for Cursor using mem0.ai - remembers context across conversations.

Features

• Auto-Save Every Prompt: Each prompt is automatically saved to mem0
• Automatic Memory Retrieval: Relevant memories are injected before each prompt
• Conversation Storage: Full conversations saved when sessions end
• Semantic Search: Vector-based search for intelligent memory retrieval

Installation

1. Copy hooks to your project

cp -r .cursor /path/to/your/project/

Or for global installation:

cp -r .cursor ~/.cursor/

2. Install dependency

pip install mem0ai

3. Configure environment

Create a .env file in your project or globally in ~/.cursor/.env:

MEM0_API_KEY=your-api-key-here
MEM0_USER_ID=cursor-user

# Optional
MEM0_TOP_K=5
MEM0_THRESHOLD=0.3
MEM0_AUTO_SAVE=true

Priority: Project .env overrides global ~/.cursor/.env

Get your API key from app.mem0.ai

4. Restart Cursor

The hooks will be active after restart.

Structure

.cursor/
├── hooks.json           # Hook configuration
└── hooks/
    ├── memory_retrieve.py   # Memory retrieval + auto-save
    └── memory_save.py       # Session-end save

How It Works

beforeSubmitPrompt Hook

1. Searches mem0 for relevant memories
2. Injects them as context
3. Auto-saves the prompt to mem0

stop Hook

1. Extracts recent messages from conversation
2. Saves them to mem0 for future retrieval

Configuration

Variable	Description	Default
MEM0_API_KEY	Your mem0 API key (required)	-
MEM0_USER_ID	User identifier	cursor-user
MEM0_TOP_K	Memories to retrieve	5
MEM0_THRESHOLD	Similarity threshold	0.3
MEM0_AUTO_SAVE	Auto-save prompts	true
MEM0_SAVE_MESSAGES	Messages to save on stop	10

Testing

Run the test suite:

pip install pytest
python -m pytest tests/ -v

License

MIT

Also Available

• Claude Code version