Does get_config expose my API key?

No. The API key is always masked in the response, so your AI can confirm a key is present without ever seeing or transmitting the actual secret.

Does get_config change any settings?

No. It is a read-only tool. It returns your current LMCP configuration and never modifies anything. Use the LMCP tray app or config file to change settings.

Why would I ask my AI to read my LMCP config?

It's the quickest way to verify your setup — network mode, port, license status, and whether an API key is set — without opening a terminal or editing files. It pairs well with run_diagnostics for self-checks.

Check LMCP Config from AI on Mac

The get_config tool returns your current LMCP configuration so your AI assistant can confirm how Local MCP is set up on your Mac. Sensitive values are protected — the API key is always masked — so you get a safe, readable view of what's active without exposing secrets. It's a read-only tool: it inspects settings and changes nothing.

A typical prompt looks like this: "Show me my current LMCP config and confirm which settings are active." Your AI calls get_config and reports back the live configuration — network mode, port, license status, and a masked API key — straight from your machine.

This is the fastest way to answer questions like "is my API key set?", "what mode is LMCP running in?", or "is everything configured correctly?" without opening a terminal or digging through config files. Download LMCP to get started — it's free and native to macOS.

Which AI agents work?

LMCP exposes 150+ tools, including get_config, to a wide range of AI clients. Desktop clients auto-configure over a local stdio connection, while web AIs connect through the LMCP Cloud Relay connector.

Claude Desktop — auto-configures via local stdio
Cursor — auto-configures via local stdio
VS Code (GitHub Copilot) — auto-configures via local stdio
Windsurf — auto-configures via local stdio
Zed — auto-configures via local stdio
Claude.ai (web) — connects through the LMCP Cloud Relay connector
ChatGPT — connects through the LMCP Cloud Relay connector
Grok — connects through the LMCP Cloud Relay connector
Perplexity — connects through the LMCP Cloud Relay connector

Automation

get_config is the natural first step in a diagnostic chain. Pair it with run_diagnostics and lmcp_state and your AI can self-verify the whole setup: "Read my config, run diagnostics, and tell me if anything needs fixing." When a tool fails, the AI can call get_config to confirm the right network mode and license status before retrying. Because it's read-only and instant, it slots cleanly into multi-step workflows without side effects.

Context

get_config reflects the real state of LMCP on your Mac — not a generic template. Your AI sees the exact settings your machine is running: the actual port, the actual network mode, the actual license status, and whether an API key is present (masked). That grounded context means the AI's troubleshooting advice is specific to your install, not boilerplate guesses.

Productivity

Instead of quitting your AI conversation to open Terminal, locate config.json, and parse it by eye, you just ask. get_config turns a two-minute file hunt into a one-line question. For anyone setting up LMCP across multiple machines, it's a quick sanity check before relying on the more powerful Mail, Calendar, and Contacts tools — confirming the foundation is solid in seconds.

Privacy & GDPR

Everything happens locally on your Mac. get_config reads your on-device configuration and returns it to your AI client over a local connection — nothing is uploaded, stored, or copied to any server. The API key is masked even in that local response, so secrets never leave your control. This is privacy by architecture, not policy: there is no cloud copy of your data to leak. Learn more about how LMCP is GDPR-compliant by architecture.