Cursor Integration Guide

{
  "mcpServers": {
    "engram": {
      "type": "http",
      "url": "https://mcp.getengram.app/mcp",
      "headers": {
        "Authorization": "Bearer engram_sk_live_your_key_here"
      }
    }
  }
}

## Engram Memory

You have access to Engram memory tools via MCP. Use them automatically.

### On session start

Search Engram for context relevant to the current task:

  search
    query: "<summary of what the user is asking about>"
    limit: 5

Include relevant results in your working context.

### During the session

Store important context:

  create_conversation
    title: "<concise description>"
    agent_id: "cursor"
    tags: ["<project-name>", "<topic>"]

  append_messages
    conversation_id: "<id>"
    messages:
      - role: "user"
        content: "<what the user asked>"
      - role: "assistant"
        content: "<what you did and why>"

### What to store

- Decisions and their reasoning
- Bug investigations and resolutions
- User preferences and coding style
- Architecture discussions

### What NOT to store

- Routine code generation
- File reads and searches
- Information already in git

You: We need to add rate limiting to the API. Use a sliding window
     algorithm with Redis, not the token bucket approach.

Cursor: [searches Engram — no prior context for this]
Cursor: I'll implement sliding window rate limiting with Redis...
Cursor: [stores in Engram:
  title: "Rate limiting: sliding window with Redis"
  tags: ["api", "rate-limiting", "architecture"]
  messages: decision to use sliding window over token bucket, Redis as backend]

You: Add rate limiting to the new /upload endpoint.

Cursor: [searches Engram → finds rate limiting decision]
Cursor: I'll add sliding window rate limiting using Redis, consistent
        with the approach we chose for the other API endpoints.
        Implementing now...