Skip to content

Getting Started

Get engram running in under 5 minutes.

Prerequisites

  • Python 3.11+
  • GEMINI_API_KEY — for LLM reasoning and embeddings (get one free)
  • Basic storage (episodic + semantic graph) works without the API key

Install

pip install engram-mem

Or from source:

git clone https://github.com/docaohieu2808/Engram-Mem.git
cd engram && pip install -e .

Initialize

engram init
export GEMINI_API_KEY="your-key"

This creates ~/.engram/config.yaml with sensible defaults.

First Commands

# Start the background daemon (HTTP server + session watcher)
engram start

# Store your first memory
engram remember "Deployed v2.1 to production at 14:00 - caused 503 spike"

# Search memories
engram recall "production incidents"

# Reason across all memory
engram think "What deployment issues have we had?"

# Check system status
engram status

Connect Your AI Agent

Add to ~/.claude.json:

{
  "mcpServers": {
    "engram": {
      "command": "engram-mcp",
      "env": { "GEMINI_API_KEY": "your-key" }
    }
  }
}

Add to Cursor's MCP settings:

{
  "mcpServers": {
    "engram": {
      "command": "engram-mcp",
      "env": { "GEMINI_API_KEY": "your-key" }
    }
  }
}

engram setup

The wizard auto-detects installed agents (Claude Code, Cursor, OpenClaw, Windsurf, etc.) and configures them interactively.

Next Steps