← Back to UniFi MCP

Privacy Policy

Last updated: March 24, 2026

UniFi MCP does not collect, store, or transmit your personal data. All communication stays between your AI agent and your local UniFi controller. We have no servers, no analytics, no telemetry, and no tracking.

How UniFi MCP Works

UniFi MCP is a set of Model Context Protocol servers that run locally on your machine. They connect directly to your UniFi controller on your local network. Your AI agent (Claude Code, Claude Cowork, or other MCP clients) communicates with the MCP server over a local stdio or HTTP connection.

Your AI Agent <--local--> MCP Server (on your machine) <--your network--> UniFi Controller

What We Don't Collect

Data Flow

Local Mode (default)

All data stays on your machine and your local network:

  1. Your AI agent sends a tool call to the MCP server (running locally)
  2. The MCP server queries your UniFi controller over your local network (HTTPS)
  3. The controller responds with the requested data
  4. The MCP server formats and returns the result to your AI agent

No data leaves your network.

Relay Mode (optional)

If you choose to enable the cloud relay, your tool calls are proxied through a Cloudflare Worker:

Third-Party Services

ServiceWhen UsedWhat It Sees
Your UniFi ControllerAlwaysTool requests over your local network
Cloudflare WorkersOnly if you enable relayTool calls in transit (encrypted, not stored)
PyPIDuring installationStandard package download
GitHubDuring plugin installationPlugin source code download

We do not integrate with any analytics, advertising, or data broker services.

Your Controller Credentials

Data Retention

UniFi MCP stores nothing. There is no database, no cache, no log files, and no session state. All data lives on your UniFi controller. When the MCP server process stops, no data persists.

Changes to This Policy

If we change this privacy policy, we will update the date at the top and commit the change to the repository. Since UniFi MCP has no data collection, changes would only reflect new features or clarifications.

Contact

For privacy questions, reach the maintainer via the GitHub repository or the email listed on the maintainer's profile.