Request a Feature from Your AI on Mac

The request_feature tool lets you tell the LMCP team what you wish your AI assistant could do — a brand-new capability, a tool that doesn't exist yet, or an app or integration you'd love to see supported. Instead of hunting for a contact form, you just ask your AI in plain language and it files the request for you.

A concrete example you can type right now: "Request a feature: add a tool that exports my Apple Notes folders to Markdown files in a folder I pick." Your AI captures the idea, structures it, and submits it to the LMCP team through request_feature — no email, no GitHub account, no leaving the chat.

LMCP is a free, native macOS MCP server with 150+ tools that let AI assistants use your real Mac apps locally. Download LMCP to get started.

Which AI agents work?

request_feature works across every major AI client that speaks the Model Context Protocol:

• Desktop clients — Claude Desktop, Cursor, VS Code (GitHub Copilot), Windsurf and Zed auto-configure through a local stdio connection. LMCP wires itself in and the tool appears automatically.
• Web AIs — ChatGPT, Claude.ai (web), Grok and Perplexity connect through the LMCP Cloud Relay connector, so you can file feature requests even from a browser tab.

Whichever client you use, the request lands in the same place: the LMCP team's backlog.

Automation

request_feature is most powerful when chained with other LMCP tools. Hit a wall mid-workflow — say your AI tried to read a calendar attachment and the capability isn't there yet — and you can immediately follow up with "file that as a feature request" without breaking your flow. Your AI can summarize what it just attempted, what was missing, and submit a precise, context-rich request in one step. It can also batch several wishes from a single conversation into separate, well-described requests.

Context

Because LMCP runs on your Mac, your AI already sees your real data — your Mail, Calendar, Contacts, Notes, Reminders, Messages, Finder files and more. That means when you describe a missing feature, the AI understands the actual apps and workflows you use. A request like "I want this to also cover my work calendar" carries real meaning, because the assistant knows which calendars you have. The result is feature requests grounded in your genuine day-to-day setup, not vague guesses.

Productivity

Filing feedback usually means stopping, switching apps, finding a form, and re-explaining context you just had in your head. request_feature collapses that into a single sentence. You stay in the conversation, the AI does the writing, and a clear, actionable request reaches the team in seconds instead of minutes. Over time, that low friction means your real needs actually get heard — and shipped — instead of being forgotten the moment you close the chat window.

Privacy & GDPR

Everything LMCP does runs locally on your Mac. There are no API keys to paste, and your apps, files and personal data are never copied to a server. When you submit a feature request, only the description you choose to send leaves your machine — your underlying Mail, Calendar and Contacts stay put. This is privacy by design: GDPR-compliant by architecture, because there's no cloud copy of your data to leak in the first place.