Cursor IDE Integration

Add persistent, encrypted memory to Cursor's AI assistant via the Model Context Protocol. Your coding context survives across sessions — securely.

On this page

Overview
Setup
Usage
Project vs Global

Overview

Cursor IDE supports MCP servers, allowing its built-in AI assistant to interact with external tools. By connecting Synapse Layer, Cursor's agent gains the ability to store and recall memories across coding sessions — project preferences, architecture decisions, debugging context — all encrypted with AES-256-GCM.

Setup

Cursor supports MCP configuration via a JSON file. Create or edit .cursor/mcp.json in your project root (or the global Cursor config directory):

json

{
  "mcpServers": {
    "synapse-layer": {
      "command": "uvx",
      "args": [
        "--from",
        "synapse-layer",
        "synapse-layer"
      ]
    }
  }
}

Note

Requires Python 3.10+ and uvx (from the uv package manager). Install uv: curl -LsSf https://astral.sh/uv/install.sh | sh

After saving the config, restart Cursor. The Synapse Layer tools will appear in Cursor's MCP tools panel.

Usage

Once connected, you can ask Cursor's AI to use Synapse Layer naturally:

You:"Remember that this project uses a monorepo with pnpm workspaces and the API is in /packages/api."

Cursor AI:Stored. I'll remember your project structure for future sessions.

Next session, different file...

You:"Where is the API code in this project?"

Cursor AI:Based on my memory, this project uses a monorepo with pnpm workspaces and the API is in /packages/api.

Project vs Global Configuration

Scope	Config Location	Use Case
Project	.cursor/mcp.json	Memory scoped to a specific codebase. Commit to version control for team sharing.
Global	~/.cursor/mcp.json	Memory available across all projects. Personal preferences and cross-project context.

Tip

Use different SYNAPSE_AGENT_ID values per project to keep memory namespaces isolated — or use the same ID for cross-project memory sharing.

Next Steps

CrewAI

Multi-agent memory with CrewAI.

LangChain

Conversational memory for LangChain agents.

MCP Overview

How Synapse Layer works as an MCP server.