mcp

@rawcontext/engram-mcp

Long-term memory for AI agents.

Overview

Engram gives AI agents persistent memory across sessions. Store decisions, preferences, insights, and context that your AI assistant can recall in future conversations.

Installation

Claude Code

claude mcp add engram -- npx -y @rawcontext/engram-mcp

Gemini CLI

gemini mcp add engram -- npx -y @rawcontext/engram-mcp

Codex CLI

codex mcp add engram -- npx -y @rawcontext/engram-mcp

OpenCode

Add to ~/.config/opencode/opencode.json:

{
  "mcp": {
    "engram": {
      "command": "npx",
      "args": ["-y", "@rawcontext/engram-mcp"]
    }
  }
}

Antigravity

Click Agent Session → ⋯ → MCP Servers → Manage MCP Servers → View raw config and add:

{
  "engram": {
    "command": "npx",
    "args": ["-y", "@rawcontext/engram-mcp"]
  }
}

Cursor

Or add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "engram": {
      "command": "npx",
      "args": ["-y", "@rawcontext/engram-mcp"]
    }
  }
}

VS Code

Or add to .vscode/mcp.json:

{
  "servers": {
    "engram": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@rawcontext/engram-mcp"]
    }
  }
}

JetBrains (WebStorm, IntelliJ, PyCharm, etc.)

Go to Settings → Tools → AI Assistant → Model Context Protocol (MCP), click Add, and configure:

{
  "mcpServers": {
    "engram": {
      "command": "npx",
      "args": ["-y", "@rawcontext/engram-mcp"]
    }
  }
}

Visual Studio

Or open GitHub Copilot Chat, select Agent mode, click the tools icon, then + to add:

• Server ID: engram
• Type: stdio
• Command: npx
• Args: -y @rawcontext/engram-mcp

Manual Configuration

Add to your MCP config file:

{
  "mcpServers": {
    "engram": {
      "command": "npx",
      "args": ["-y", "@rawcontext/engram-mcp"]
    }
  }
}

On first run, authenticate via browser.

Tools

Tool	Description
remember	Store memories with categorization (decision, insight, preference, fact)
recall	Search memories with semantic and keyword matching
context	Assemble comprehensive context for a task
query	Run custom queries against your memory graph
summarize	Condense text into key points
extract_facts	Parse text into atomic facts
enrich_memory	Auto-generate summary, keywords, and category

Resources

URI	Description
memory://{id}	Individual memory by ID
session://{id}/transcript	Full conversation transcript
session://{id}/summary	Session summary
file-history://{path}	Change history for a file

Prompts

Prompt	Description
/engram:session-prime	Load context for a new task
/engram:session-recap	Summarize a past session
/engram:decision-history	Investigate past decisions on a topic

License

AGPL-3.0