Launched this week
LUCI Desktop
Let your AI agents remember what you've seen
817 followers
Let your AI agents remember what you've seen
817 followers
Luci saves your screen history and meeting transcripts locally, so agents like Claude Code, Cursor and Codex can help you find a page you forgot to bookmark or recall a decision from a call. Capture context across apps without setting up a connector for each one. Includes on-device transcription and daily summaries with Microsoft Foundry Local. Free for Mac and Windows.
Free
Launch Team
Context.devAPI to scrape anything for AI, 20,000+ developers trust us.
Promoted
Have been looking for a better option ever since Rewind was taken down. This seems to be super promising. However, it looks like registration is full and waitlisted.
just tested it on mac to find a doc link I forgot to bookmark earlier pulled it up in seconds.
Love the no connectors approach, especially with everything staying local. Congrats!
@henry_habib Thank you! We wanted it to work across the apps you already use, without having to connect each one. Glad that resonated!
@henry_habib Thanks Henry. Connectors break, need permissions, and only cover the apps someone built one for. The screen is already the common layer, so we started there.
What does the setup process look like for connecting LUCI to an AI agent?
Congrats on the launch! Question from a heavy Claude Code user: I usually have two sessions running side by side on different parts of the same project. Can both query Luci at the same time without stepping on each other? And more interesting for me, would session B be able to recall what session A was doing on screen, so Luci works as shared context between them?
@yevhen_skrypets Great question, and this is exactly how I run it myself. Luci runs a single local MCP server on your machine over HTTP, so any number of Claude Code sessions can connect and query it at the same time without stepping on each other. And yes to the second part: Luci captures what is on your screen, so if session A's work was visible, session B can ask what A was doing and pick up from there. There is no per-session sandbox, the whole screen history is one shared memory. Would love to hear how the two-session setup feels once you try it.
This would genuinely help with how scattered my day gets across tabs and calls. Does it handle multi-monitor setups the same way, or does capture only track one screen at a time?