Claude Overlay
A free, screen-aware Claude Code chat for Windows that floats over your work, reads every monitor, and can edit the file you're looking at.
If you're on Windows and already paying for Claude, Claude Overlay removes the most annoying step in using an AI assistant — describing your screen. Multi-monitor capture on every query, handed to a real Claude Code agent that can edit the open file, is the whole product, and it's free on top of a subscription you own. Cheaper than paying for a separate API tier: it runs on your own claude CLI login. The tradeoffs are setup friction (Python and CLI auth) and a hard Windows-only wall, both documented in the README. If you want a zero-setup chatbot, use the Claude desktop app instead.
Verified 10h ago · liveness 64/100 · cite: rightaichoice.com/tools/claude-overlay
- Windows developers debugging an error dialog or stack trace they can't easily copy
- Power users editing an open PowerPoint or Excel file with agent-driven automation
- Analysts building a spreadsheet with live formulas already open in Excel
- Anyone on multiple monitors who wants the assistant to see all of them at once
- macOS or Linux users — the README lists Windows 10/11 only
- People who won't touch a terminal — setup needs Python and CLI authentication
- Anyone expecting zero-setup chat like the Claude desktop app
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Claude Overlay if you're on macOS or Linux, or you want a zero-setup chatbot you can install and use without touching Python or Claude CLI authentication.
The app itself is free, but every message runs against your existing Claude subscription — heavy screen-reading use eats into the same usage limits as your other Claude work.
Free, MIT-licensed, and it costs nothing beyond the Claude subscription you already pay for. Per-seat AI assistants like the Claude desktop app are included with a subscription too, but tools that charge for screen context or agent execution typically add $20–$30/mo per seat. Claude Overlay's fit is individual Windows power users and small teams on Claude Pro or Max; there's no team plan, SSO, or admin console because there's no paid tier at all.
In short
Claude Overlay — A free, screen-aware Claude Code chat for Windows that floats over your work, reads every monitor, and can edit the file you're looking at. Best for Windows developers debugging an error dialog or stack trace they can't easily copy, Power users editing an open PowerPoint or Excel file with agent-driven automation, Analysts building a spreadsheet with live formulas already open in Excel. Free to use.
What people actually say about Claude Overlay — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
61 mentions across 4 sources (YouTube, Product Hunt, Bluesky, Lemmy) · researched Jul 16, 2026.
Average across the 4 sources that answered — each source counts once, not each post.
- +Eliminates alt-tabbing and manual screenshot pasting for Claude users.
- +Uses existing Claude CLI login—no extra API key needed.
- +Multi-monitor support automatically captures each screen.
- +Voice input paired with screen sharing adds hands-free versatility.
- +Polished hotkey activation (Ctrl+Alt+Space) feels seamless.
- −No option to mask sensitive screen regions before capture.
- −Prompt injection from on-screen text is a real security risk.
- −Token costs can escalate quickly with frequent screenshots.
- −Unclear if overlay shares rate limits with terminal Claude sessions.
- −Accidentally closing the overlay can delete unsent input.
- • Screenshot-heavy usage can increase token consumption and cost if on a pay-per-use plan.
Viability Score
How well maintained and how widely used is Claude Overlay? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: September 2026
How we score →Key Features
- Screen-aware chat that auto-captures every monitor on each message
- Labels primary vs. secondary monitors in the captured context
- Full Claude Code agent (Opus 4.8) that edits files and runs shell commands
- Drives your open app (PowerPoint, Excel) via PowerShell/COM automation
- Edits the file you're looking at without needing its path
- Frameless always-on-top floating window over any application
- Collapses to a draggable orb with a real Windows taskbar button
- Checkmark on the orb when a reply finishes while collapsed
- Name each overlay instance and run several at once
- Token-by-token streaming responses with tool-call chips
- In-place model switcher and context-usage meter
- Ctrl+V image paste for asking about a pasted screenshot
- Skips re-sending the screen when it hasn't changed since last message
- One-click resume of the last conversation after a restart
- Reconnects into the same conversation if the CLI connection drops
About Claude Overlay
Claude Overlay is a frameless, always-on-top chat window for Windows 10 and 11 that floats over every app you already have open. Ask a question and the full Claude Code agent (Opus 4.8, per the README) looks at your real screen — auto-capturing each monitor on every message and labelling primary versus secondary — so "what's wrong here?" is a complete question with no copy-pasting, no describing, no alt-tabbing. Because it runs a real agent rather than a chatbot, it reads, edits, and runs things where you work: the README's example has it drive PowerPoint through PowerShell/COM automation to fix wording on the open slide, without you supplying a file path. It's MIT-licensed and free — it drives your own claude CLI login, so it uses your existing Claude subscription with no API key and no metered billing. Day-to-day ergonomics are more considered than most hobby overlays: it collapses to a draggable orb with a real Windows taskbar button, drops a ✓ on the orb when a tucked-away reply finishes, and lets you name each instance so a row of orbs stays tellable apart. Responses stream token-by-token with tool-call chips, plus an in-place model switcher and a context-usage meter. Conversations survive an app restart through one-click resume, and a dropped CLI connection reconnects into the same session. Setup is not download-and-go: it wants Python and CLI authentication. Compared with the Claude desktop app, the overlay adds screen reading plus agent execution; compared with a terminal Claude Code session, it adds the screen.
Behind the Verdict
Claude Overlay's bet is narrow and correct: the hardest part of asking an AI about your work is the description step, and a Windows overlay that already sees your screen deletes it. The README is unusually concrete about how — every message auto-captures each monitor and labels primary versus secondary, and if your screen hasn't changed since the last message the duplicate screenshot isn't re-sent. That second detail is the kind of thing hobby overlays usually skip. The bigger differentiator is that it isn't a chat wrapper. It runs the full Claude Code agent (Opus 4.8), so it edits files and runs commands — the README's example is fixing the wording on an already-open slide by driving PowerPoint through PowerShell/COM automation, on the file you're looking at, with no path required. [inferred] That matters for people whose work lives in Office apps rather than in a repo. Ergonomics are considered. It collapses to a small draggable orb with a real taskbar button to bring it back, marks the orb with a ✓ when a tucked-away reply finishes, and lets you name each instance so several orbs stay distinguishable. Responses stream token-by-token with tool-call chips and a context meter; there's an in-place model switcher. Conversations survive restart through one-click resume, and a dropped CLI connection reconnects into the same session. Where it doesn't fit: it's Windows 10/11 only, wants Python plus CLI authentication, and the README itself warns the app-driving approach is best-effort, so don't point it at business-critical documents unattended. Compared with the Claude desktop app, the overlay adds screen reading plus agent execution. Compared with a terminal Claude Code session, it adds the screen. Pick the one that matches where your work actually happens.
Researching Claude Overlay? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Claude Overlay actually fits — and what changes day-one when you adopt it.
An exception throws in an app you're testing and the stack trace is in a dialog you can't easily copy. You summon the overlay, which already captured every monitor, and type 'what's wrong here?'
Outcome: The full Claude Code agent reads the trace from the screen, explains the failure, and can edit the offending file directly — no alt-tab to a browser, no manual paste.
You have a spreadsheet open with live formulas and want a model built or a formula fixed. You ask the overlay while the workbook stays in front, and it reaches into Excel via PowerShell/COM automation.
Outcome: The agent acts on the file you're looking at without you supplying its path, so the workbook changes in place while you keep your flow.
You name one overlay per task and run several at once. When you collapse them, each orb carries its task name and shows a ✓ when a tucked-away reply finishes.
Outcome: A row of orbs stays tellable apart, you know which task is done without expanding anything, and conversations survive a restart through one-click resume.
Use Cases
- Debug code by capturing error messages on screen and getting instant explanations
- Summarize long articles or documents visible on screen without manual typing
- Translate foreign language text displayed in any application in real time
- Generate code snippets based on UI elements or data tables visible on screen
- Compose emails or messages by dictating thoughts while seeing the context
- Learn concepts by asking questions about diagrams or charts currently displayed
Models Under the Hood
as of 2026-09-09
Limitations
- Windows 10/11 only; no macOS/Linux support.
- Setup requires manual command-line steps (Python, Claude CLI auth).
- The app-driving approach (PowerPoint/Excel via PowerShell/COM) is documented as best-effort, so don't rely on it for unattended work on critical documents.
- It depends on your own claude CLI login and a working connection to the CLI; usage counts against your existing Claude subscription limits rather than a separate API bill.
- It is not a plug-in chatbot — if you want zero-setup chat, use the Claude desktop app instead.
as of 2026-09-15
Verification history
We have re-verified Claude Overlay 6 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-checked, vendor evidence unchanged
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
Free to cite with attribution — this page re-verifies continuously.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Plans compared
For each published Claude Overlay tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Free (open source)
$0
Ideal for
Individual Windows 10/11 power users and small teams who already pay for Claude Pro or Max and want screen context without a separate API bill.
What this tier adds
Starting tier and the only tier: MIT-licensed at $0, running on your own claude CLI login with multi-monitor capture and the full Claude Code agent.
Where the pricing makes sense
The company stage and team size where Claude Overlay's pricing actually pencils out — and where peers do it cheaper.
Free, MIT-licensed, and it costs nothing beyond the Claude subscription you already pay for. Per-seat AI assistants like the Claude desktop app are included with a subscription too, but tools that charge for screen context or agent execution typically add $20–$30/mo per seat. Claude Overlay's fit is individual Windows power users and small teams on Claude Pro or Max; there's no team plan, SSO, or admin console because there's no paid tier at all.
Setup time & first value
How long it actually takes to get something useful out of Claude Overlay — broken out by persona, not the marketing-page minute.
Not download-and-go for anyone: the README expects Windows 10/11, an installed Python, and authentication against your own claude CLI login. The repo ships setup.cmd, install-python.ps1, Create Desktop Shortcut.cmd and Diagnose.cmd, so a comfortable Windows user can plausibly be talking to it in 15–30 minutes; if you've never touched a terminal, budget an hour plus troubleshooting. First value
Switching to or from Claude Overlay
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From terminal Claude Code: keep your existing claude CLI login, then run the overlay on top so the same agent gets screen context it never had in a shell.
- →From the Claude desktop app: install Python plus the repo, authenticate the CLI, and you gain multi-monitor screen reading and agent file edits the desktop chat doesn't do.
- →From a manual copy-paste workflow: point the overlay at the app you're already in and stop describing your screen — it captures every monitor on each message.
- ↗To terminal Claude Code: your login is already the same, so drop the overlay and run the CLI directly if you don't need screen context.
- ↗To the Claude desktop app: use it when you want zero-setup chat and don't need Windows-only screen capture or app-driving automation.
Resources & Guides
- Resourcegithub.com
SETUP · Claude Overlay
Helpful link from github.com
- Resourcegithub.com
README · Claude Overlay
Helpful link from github.com
- Resourcegithub.com
CONTRIBUTING · Claude Overlay
Helpful link from github.com
- Resourcegithub.com
CHANGELOG · Claude Overlay
Helpful link from github.com
- Resourcegithub.com
Claude Overlay · Claude Overlay
Helpful link from github.com
Tutorials & Learning
YouTube returned 6 videos for “Claude Overlay”, and we withheld 6: 6 did not mention Claude Overlay. We are showing none, because we could not prove any of them are about Claude Overlay.
Official links
Tools that pair well with Claude Overlay
Common stack mates teams adopt alongside Claude Overlay, with the specific reason each pairing earns its keep.
Featured Head-to-Head Comparisons
Claude Overlay vs Cognition Ai
If you're a solo developer or power user who wants AI to see and act on your screen without leaving your workflow, Claude Overlay is your lightweight, free overlay for Windows. But if you're leading an enterprise team that needs an autonomous engineer to plan, code, test, and ship PRs (with a financial guarantee), Cognition AI's Devin is the heavy lifter with proven ROI at Fortune 500 companies. Choose based on scale: individual productivity vs. organizational transformation.
Claude Overlay vs Poke Interaction Co
If you're a Windows power user who wants an AI agent that directly interacts with your screen across multiple monitors, Claude Overlay is your tool. If you prefer managing email, calendar, and tasks through your existing messaging apps with rich integrations like Notion and Oura, pick Poke. Both are freemium but serve fundamentally different workflows: desktop automation vs. mobile-first assistant.
Claude Overlay vs Gem
Gem and Claude Overlay serve entirely different purposes: Gem is a full-stack recruiting platform for talent teams, while Claude Overlay is a screen-aware AI assistant for developers. Your choice depends on whether you need to streamline hiring (Gem) or boost coding/ productivity with always-on AI (Claude Overlay). They are not substitutes.
Alternatives to Claude Overlay
View allGemini
Google's native AI assistant for Workspace, with multimodal input and computer-use automation
PromptDrive.ai
Organize, share, and collaborate on ChatGPT, Claude, and Gemini prompts in one workspace.
Frequently Asked Questions
Best-of guides
Used Claude Overlay? Help shape our editorial sentiment research.