Launching today
Claude Overlay
A floating Claude Code chat that sees your screen
107 followers
A floating Claude Code chat that sees your screen
107 followers
Claude Overlay is a frameless, always-on-top chat window for Claude Code that floats over everything you do. Summon it with a hotkey, ask about whatever's on screen, and Claude captures and reads your monitors before answering. It runs the full Claude Code agent on your own subscription - no API key - so it can edit files and run commands, not just chat. Open source (MIT), Windows.
Love how it pulls from your actual screen context, that's the killer feature for me. One thing that would push it further: a quick toggle to switch between which monitor Claude reads, since I often have a reference doc on my second display I don't want it touching.
@pakizecankara Thanks — that "reads your actual screen context" part is the whole reason it exists, so I'm really glad it clicks for you. And you're the second person to ask for this, which tells me it's clearly the next thing to build: right now it grabs every monitor at once, so your second-display reference doc gets pulled in whether you want it or not. A quick per-monitor toggle — "only read this screen" — is squarely on the roadmap. In the meantime you can just tell Claude "only look at my main display" and it'll focus there (each screen is labelled in the prompt), but a real toggle is the proper fix. Appreciate it
Really like that this sits on top of the existing Claude Code login instead of adding another agent layer.
One thing I’d want for daily coding use is a small run receipt after each screen-driven action: screenshot source, files touched, commands run, and which step came from what was visible on screen. When the input is a screen, debugging a wrong action later gets harder than with a normal prompt.
That kind of trace would make the overlay feel safer for long work sessions, not just faster.
@grace_lee26 Thanks — "on top of your existing login, not another agent layer" was a deliberate choice, so I'm really glad that lands.
And this is a great ask — honestly the missing half of the safety story a few people have raised in this thread. You're right that screen input makes provenance murkier: with a text prompt you can re-read exactly what you asked, but "what was on screen" is fuzzier to reconstruct later.
Where it stands today: every command and file edit already streams into the chat as a visible tool chip as it happens, and each turn notes which monitors were attached — so there's a live trace, just not a consolidated, reviewable receipt. Under the hood the captured frames are saved to a temp folder and there's a debug log, so the raw material for a proper trace already exists.
What you're describing — a compact per-action receipt (screenshot source → commands run → files touched) you can scan after the fact — is very buildable and I want it. Honest nuance: the coarse version (this turn captured monitors 1+2, ran X, edited Y) is straightforward; the fine-grained "this specific edit traces to this region of the screen" is harder to do reliably, so I'd likely ship the per-turn receipt first. It pairs naturally with the read-only mode a couple of others asked for — trace + scoped permissions is exactly what makes it feel safe for long sessions, not just fast. Going on the roadmap; MIT if you'd ever want to help shape it. Excellent suggestion
Since you're using the full Claude Code agent instead of just an API, what's the most surprising workflow users have discovered? Has anyone been using it for something completely different than debugging?
finally something that lets claude see my actual screen instead of me copy-pasting everything. the always-on-top window is genuinely handy when you're jumping between docs and code.
@busefk6b Thank you — that copy-paste loop was the exact itch I built this to scratch. The "always-on-top + just ask about whatever's on screen" combo is the whole point, so it's really good to hear it lands that way when you're bouncing between docs and code. If you end up wanting anything (more shortcuts, cross-platform down the line), I'm all ears 🙏
Always-on-top framing feels really natural when I'm jumping between editor and browser. The fact that it reads my actual screen and runs as a real agent on my plan makes it way more useful than another chat sidebar.
@pelinokutuaden Thank you — "more useful than another chat sidebar" is exactly the bar I was aiming for. The always-on-top + reads-your-actual-screen + real-agent-on-your-own-plan combo is the whole thesis: not a chat box bolted on the side, but something that's already in the loop while you're bouncing between editor and browser. Really glad it feels that way in practice. If you ever hit a rough edge or want something added, I'm all ears
Life saver!