Every conversation with an AI coding assistant starts the same way: you explain what you are working on, which files you changed, what your teammate said in Slack, and which ticket you are on. Lurk, a new open-source macOS tool, wants to eliminate that ritual entirely.
Lurk runs as a local daemon that quietly watches what you are doing on your machine and makes that context available to AI tools like Claude Code, Cursor, and ChatGPT. The idea is straightforward: instead of manually pasting context into every prompt, Lurk builds a rolling picture of your work session and hands it to your AI assistant automatically.
How It Works
Lurk sits in the background on macOS and tracks activity across your development environment. When you switch to an AI tool, it can supply context about your recent file changes, the conversations you have had in other apps, and the general flow of your work session. It connects via MCP (Model Context Protocol), the same standard that Claude Code and other tools already support for pulling in external data.
The key difference from existing MCP connectors is scope. Slack integrations give your AI assistant Slack messages. GitHub integrations give it repo data. Lurk tries to give it everything at once, the same ambient awareness a human colleague would have if they were watching your screen.
Privacy Tradeoffs
All processing happens locally, which is the right call for a tool that literally watches your screen. No data leaves your machine unless you explicitly send it to an AI provider as part of a prompt. Still, running a daemon that monitors your activity is a trust decision. The project is open source, so you can audit the code, but you should understand what it captures before installing it.
Lurk is macOS-only for now and available on GitHub. If you spend half your day re-explaining context to AI assistants, it is worth a look.