Install
Get it from the Chrome Web Store.
One click, and Chrome keeps it up to date. No zip to keep, no folder to remember, no Developer mode.
Add to Chrome
Open the Pagehand listing and click Add to Chrome.
Pin it
Click the puzzle icon in the toolbar and pin Pagehand, so the side panel is always one click away.
Open the side panel
Click the Pagehand icon on any page. Sign in — or switch to your own API key — and start asking.
Updates arrive on their own. Prefer building it yourself? See the GitHub setup.
Rather load it unpacked?
Download
Get pagehand.zip from the latest GitHub Release.
Unzip
Extract the zip to a folder you will keep — Chrome reads that folder every time it starts. Don’t delete it after installing.
Open Extensions
In Chrome’s address bar, go to chrome://extensions and press Enter.
Turn on Developer mode
Toggle Developer mode in the top-right corner of that page.
Load unpacked
Click Load unpacked, then choose the folder you unzipped (the one that contains
manifest.json).
To update later: download the new zip, replace the folder contents, then click the reload icon on chrome://extensions.
Model access
Sign in, or bring your own key.
Two ways to reach a model. The hands are the same either way — the agent loop and every CDP tool run inside your browser.
Hosted · default
Sign in and go
One email, no API key, no provider account. Pagehand routes the model call through its own endpoint and carries the bill.
- Nothing to configure on the first run
- Chat threads still stay in your browser
- In private preview while usage limits are built
Your key · advanced
Bring your own key
Point the panel at DeepSeek, OpenAI, Anthropic, or any OpenAI-compatible endpoint. Requests go straight from your browser to that provider.
- Key stays in
chrome.storage.local - Nothing touches Pagehand’s servers
- Any model your provider offers, at its price
- Key stays in
Hosted is in private preview: sign-in works, but requests are served to a short list of accounts until quotas land. Email us for access → Your own key works for everyone today, no account needed.
How it works
The side panel owns the debugger.
No Node sidecar. No MCP bridge. Pagehand talks Chrome DevTools Protocol from inside the extension and drives the page you are looking at.
Open the panel
⌘/Ctrl+Shift+K, or click the toolbar icon.
Sign in
One email and the model is handled for you — or switch to your own key in Settings.
Ask it to act
Summarize, click, fill forms, chase console errors, or walk a multi-step flow on the live page.
Capabilities
Same protocol Puppeteer speaks.
Accessibility snapshots, input, navigation, screenshots, console and network — implemented against CDP domains, not a brittle DOM scraper.
take_snapshotRead the page as a structured accessibility tree with stable uids.
click / fill / typeDrive forms and UI the way a user would — including select-all before replace.
navigate / pagesMove between pages and manage them while the agent keeps context.
console / networkInspect errors and requests without opening DevTools yourself.
evaluate_scriptRun focused page scripts when reading the tree is not enough.
take_screenshotCapture what the page looks like and keep it in the chat thread.
Privacy
The hands stay local either way.
Page content is read in your browser and chat threads never leave it. On hosted, prompts and page context pass through pagehand.app to the model provider — we meter what a request cost, not what it said. With your own key, nothing touches our servers at all: it stays in chrome.storage.local and goes straight to the provider. While attached, Chrome shows a “Pagehand is debugging this browser” banner — that is intentional and cannot be hidden. Read the full privacy policy →