browse
Supported Chromium browser that the extension can connect to, letting agents open, group and drive tabs in Arc, take screenshots, capture network traffic and record the screen through chrome.debugger.
Supported Chromium browser the agent can control: it can open and drive tabs, hand tabs over to the user for logins or payments, and record or capture traffic from Brave.
Supported Chromium browser in which the extension runs agent commands, enabling tab automation, screenshots, network capture and screen recording.
Supported Chromium browser the extension can drive, letting agents open and control tabs, capture traffic and record the screen.
Supported Chromium browser in which agents can open and drive their own tab groups, take screenshots, capture network traffic and record the screen via the extension.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Let your AI agents use your real browser: your tabs, your logins, your extensions. A small native MCP server and a Chromium extension, for OpenCode, Claude Code, Codex, Cursor and any other MCP client.
Agents get a browse MCP server. They open tabs in the background, read pages, click, type, take screenshots,
capture network traffic and record the screen, in tabs grouped out of your way. When it's your turn (a login, a 2FA
prompt, a payment), they hand the tab back to you and continue once you click Continue.
It works in Chrome, Edge, Brave, Arc, Helium, Vivaldi, Opera and Chromium.
Install
curl -fsSL https://opencode.ai/browse/install | bashOr paste this into your agent: "Set up browse so you can use my Chrome: run
curl -fsSL https://opencode.ai/browse/install | bash". It sets browse up for the agent that ran it.
Other ways to install, all of which end in the same browse install:
How | Command | Where |
Install script (macOS, Linux) |
|
|
Install script (Windows) |
|
|
npx, bunx |
| copied to |
npm |
| npm's global |
bun |
|
|
pnpm, yarn, volta |
| that manager's |
There's always exactly one browse: a global package install is left to its package manager (nothing is copied next
to it, and a leftover ~/.local/bin/browse from an earlier install is removed), and agents run the same browse your
terminal does. If browse install finds another browse on your PATH it tells you which one and how to remove it.
The install script reads the channel's manifest from Anomaly's release service, downloads
the binary for this computer and checks its size and SHA-256 before installing it. --channel stable|beta|dev picks
the channel (stable by default); --version <version> installs a specific version from npm, checked against npm's
published SHA-512 integrity. With npm, @opencode/browse@dev is the dev channel. The npm package is a small Node
launcher (Node 22+) that runs the native binary npm installed for your platform.
What browse install does
Puts browse in place (see above) and adds
~/.local/binto your PATH when it copied browse there (--no-modify-pathskips that).Adds the browse MCP server to your agents. In a terminal you get a picker: OpenCode and the agents found on this computer start checked (your last choice next time). Type to search all supported agents, ↑/↓ to move, Space or Tab to toggle, Enter to confirm. An agent running
browse installsets up itself. Without a terminal OpenCode is the default. Edits keep the comments and formatting of each agent's config.Connects the extension. If no browser has browse yet, install opens it in the Chrome Web Store; add it, and install pairs it once it's there. If a browser already has it, install opens its setup page with a one-time pairing code in the link, so it pairs without copying anything. Over SSH or without a display it prints a code instead.
Then ask your agent to "open news.ycombinator.com and tell me the top story". An agent that was already running needs a restart (or an MCP reload) to load browse.
browse install -a claude codex # these agents, without asking (--agent; names below)
browse install -a '*' # every supported agent, installed or not
browse install -y # every agent found on this computer
browse install --agent none # no agent: print the MCP entry to add by handCONFIG is $XDG_CONFIG_HOME or ~/.config. APP is ~/Library/Application Support on macOS, %APPDATA% on
Windows, or CONFIG on Linux. The paths below are defaults; a client's existing config or location override can apply.
Agent (flag) | User config | Format / server key |
OpenCode ( |
| JSONC · |
Claude Code ( |
| JSON · |
Claude Desktop ( |
| JSON · |
Cursor ( |
| JSON · |
Codex ( |
| TOML · |
| JSONC · | |
Windsurf ( |
| JSON · |
Amp ( |
| JSONC · literal |
Antigravity ( |
| JSON · |
Augment (Auggie CLI) ( |
| JSON · |
Cline ( |
| JSON · |
CodeBuddy Code ( |
| JSONC · |
Cortex Code ( |
| JSON · |
DeepSeek Harness ( |
| YAML · |
Factory Droid ( |
| JSON · |
Gemini CLI ( |
| JSON · |
GitHub Copilot CLI ( |
| JSON · |
Goose ( |
| YAML · |
Grok Build ( |
| TOML · |
Hermes Agent ( |
| YAML · |
Junie ( |
| JSON · |
Kilo Code ( |
| JSONC · |
Kimi Code CLI ( |
| JSON · |
Kiro ( |
| JSON · |
LM Studio ( |
| JSON · |
Mistral Vibe ( |
| TOML · |
Pi ( |
| JSON · |
Qoder CLI ( |
| JSON · |
Qwen Code ( |
| JSON · |
Roo Code (VS Code) ( |
| JSON · |
Trae CN ( |
| JSON · |
Warp ( |
| JSON · |
WorkBuddy ( |
| JSON · |
Zed ( | macOS: | JSONC · |
For any other stdio MCP client, run browse mcp:
{ "mcpServers": { "browse": { "command": "/home/you/.local/bin/browse", "args": ["mcp"] } } }(browse install --agent none prints this with your path.) Re-running install updates the entries it wrote and
leaves servers named browse that it didn't write alone.
Pairing
There's nothing to sign in to. Each browser pairs with browse once and gets a key of its own (see
Security). If pairing didn't happen by itself, run browse install again, or run browse pair: it
prints a one-time code (XXXX-XXXX, good for 10 minutes) and waits. Open Extensions → browse → Options, expand
advanced (or click Pair with a code), enter the code under Pairing code and click Connect. browse pair --list shows paired browsers; browse pair --revoke <browser> unpairs one.
Browser on another computer
When your agent runs on one machine and your browser on another (over Tailscale, say), browse runs next to the agent and the extension connects to it across the network:
On the agent's machine run
browse install --remote(orbrowse set hostname 0.0.0.0after installing). browse then listens on every network, restarting the relay if it runs, and prints this machine's address and a pairing code. Other computers reach only the extension link; the HTTP API answers on127.0.0.1only.In the browser, open browse's setup page, choose Connect an agent on another computer and enter the address and code (or use Relay address and Pairing code under advanced on the options page).
The connection isn't encrypted by browse; use a network that is (Tailscale or WireGuard), not the open internet.
browse set hostname 127.0.0.1 goes back to this computer only.
Updates
Updates are automatic. Every relay checks at launch and then hourly, whatever the update setting, and installs the channel's release when it isn't the one running (a rollback reaches everyone too):
every install reads its channel's manifest from the release service (
stable, ordev/betafor those builds), withbrowse/<channel>/<version>/cli/<install kind>as the User-Agent;the install script's binary replaces
~/.local/bin/browsewith the manifest's tarball after checking its size and SHA-256;global installs run the package manager that installed them (
npm install -g,bun add -g,pnpm add -g,yarn global add,volta install) to install@opencode/browse@<the manifest's version>into the same place. A release reaches the service a few minutes before npm, so a package manager that doesn't have the version yet is retried quietly;npx,bunx,pnpm dlxandyarn dlxruns resolve the package themselves: they only tell you another release is out.
A relay browse started restarts on the new version when it's idle (one you run with browse serve in a terminal keeps
running until you restart it); agent tabs stay open and come back. A running MCP server keeps
working with the newer relay until the agent restarts it. browse update updates now and browse update --check only
looks. Choose Tell me or Off under updates on the extension's options page, or run browse update --mode notify|off (OPENCODE_BROWSER_AUTO_UPDATE=off in the environment overrides both). Off still checks hourly, and
installs nothing. To go back, rerun the install
script with --channel <channel> or --version <version>, or install @opencode/browse@<version> with your package
manager. The Chrome Web Store updates the extension.
Dev builds: every push to dev publishes one: curl -fsSL https://opencode.ai/browse/install | bash -s -- --channel dev, or npx @opencode/browse@dev install. The store extension works with dev builds. To try the dev
extension too, download https://anoma.ly/release/browse/extension/dev/chromium, unzip it, turn on Developer mode on
your browser's extensions page, click Load unpacked, pick the folder and run browse pair (browse extension
prints these steps on the dev channel). Remove the store extension first so only one browse runs in that profile.
Uninstall
browse uninstallIt removes browse from every agent config it was added to, stops the relay, deletes ~/.local/bin/browse (for the
install script and npx installs), the PATH line it added and browse's files. A global package stays with its package
manager: uninstall prints the command (npm rm -g @opencode/browse, bun remove -g @opencode/browse, ...). Remove the
extension from your browser to finish.
Related MCP server: chrome-mcp
Using it
Tabs are the handle.
navigatewithout atabopens a new background tab and returns its id; every other call passes thattab. Agents use the tab ids they opened or were given, and leave other tabs alone unless you ask. Tabs can be opened with options that stick:group(the tab group's title, and a handle to close them together),allowedOrigins,readOnly,keepVisible,keepOpen(never close it as idle) andbrowser. Popups a page opens become agent tabs too.Tabs keep their ids, and an agent can name one.
tabswithaction: "new"and anidlikework-mailopens that tab once: asking again returns the same tab without navigating it, and every tool takeswork-mailastab. The id lasts while the tab stays open: a relay restart or extension reload keeps it, and a browser restart loses it (opening that id again is a fresh tab).tabslists each tab's id, url, title, group, last use and whether a call is running in it. A tab opened withkeepOpen: truestays until an agent or you close it; every other tab closes when idle like the rest.Agents open their own tabs in the background of the window you last used, grouped together and collapsed unless you're looking at one (you can turn that off). Agents can't switch you to their tabs or bring a window to the front. Tabs nobody touches for 10 minutes close by themselves (never while something runs on them). Closing agent tabs never quits the browser: on Windows and Linux, browse leaves a new tab (or window) behind when it would close the last one.
Relay restarts keep agent tabs. The relay saves the tabs agents opened and their options, and takes them back when it starts again (after an update, say).
Hand an agent one of your tabs by clicking the toolbar button on it. The badge shows ON; click again to take it back. While an agent works the badge reads RUN, and WAIT when it needs you. The button always works: a click while an agent runs takes the tab back. The browser's debugging bar (Cancel, or OK in some browsers) disconnects browse from every tab; browse shows a notification when that happens, and clicking the browse button on a tab lends it again. Chrome's own pages, the Web Store included, can't be lent.
Handoffs: the page shows a small card with the site's origin. Do the human part, then click Continue on the card or the toolbar button. While a tab is recorded you get a system notification instead of the card.
Several browsers at once. Each connected browser has an id (like
chrome-macbook);statuslists them, and new tabs open in the default browser unlessbrowsernames another.Settings (the extension's options page, applied right away): where agent tabs open (your window or their own), grouping (one shared group, one per task, or none), whether browse collapses agent groups by itself or leaves them as you set them (a handoff still opens its group so you can find the tab), when idle tabs close (10, 30, 60 minutes or never), whether agent tabs open muted, human-like input, and handoff notifications. Agents can read and change these with the
settingstool when you ask them to.
Tools
OpenCode, which composes tools in code mode, gets the plain tools. Every other MCP client gets execute, status and
tabs; execute runs JavaScript that can call all of them. The relay decides by the client's name
(OPENCODE_BROWSER_TOOLS=plain|execute|all on the relay overrides it).
Tool | What it does |
| Open a page: without |
| A compact outline of the page with refs ( |
| Act on a ref, selector or point, with human-like input |
| Wait for text, a URL or an element; answer dialogs; set files; run JavaScript in the page |
| The viewport, full page, an element or a rectangle; compare with a baseline |
| Capture a tab's traffic (credentials redacted) or its console messages and errors |
| Record a tab to a video file |
| Viewport, color scheme, user agent, geolocation and more; tools a page offers through WebMCP |
| List, open or close tabs ( |
| Ask you to do something only a person can do; ask you to show a flow while it records your steps |
| Read or change your settings when you ask; the relay, browsers, pairing state and what to do next |
| For other clients: JavaScript with a Playwright-style |
Everything is a CLI too: browse navigate https://example.com prints the new tab's id, then browse snapshot --tab <id>, browse click e12 --tab <id>, browse execute --file script.js. browse tools lists them and browse <tool> --help shows a tool's options.
How it works
The agent starts browse mcp, a small stdio proxy, which starts the relay on demand and forwards each call to it on
127.0.0.1:19988 (or the next free port from 19990–19997 when another program holds it). The extension connects to
the relay over a WebSocket and runs its commands in your browser through chrome.debugger.
The relay (
browse serve, inrust/) keeps agent tabs, handoffs, recordings and network captures, talks the Chrome DevTools Protocol to each tab, and runsexecutescripts in an embedded JavaScript engine (QuickJS). It runs until it's stopped (browse restart,browse uninstall, a signal) or restarts onto an update.The extension (
src/) runs the relay's commands throughchrome.debugger, keeps agent tabs in collapsed groups by default (or their own window), closes idle ones, and shows the toolbar badge and the handoff card.Quiet in the page. browse doesn't enable the CDP domains pages can detect, runs its own code in isolated worlds and leaves no globals behind. Input moves like a person: curved mouse paths, paced typing, gliding scroll (turn Human-like input off for full speed). This makes an agent hard to spot, not impossible: the browser shows its "started debugging" bar while the debugger is attached.
Permissions: debugger, tabs, tabGroups, storage, alarms, scripting (to add the handoff card only to tabs
agents use), webNavigation (to keep popups agents open with their other tabs), notifications (to tell you an agent
needs you while it records a tab), and access to all sites so agents can work wherever you ask. Nothing leaves your
computer except what your agent sends to its model provider; see the privacy policy.
Security
browse drives the browser you're signed in to, so it assumes that what an agent reads (a web page, an email) may try to steer it, and that other programs on your computer may try to use the relay.
Relay token. Every request to the relay needs the token in
~/.local/share/browse/relay-token(readable only by you). The MCP server and CLI read it for you. The relay listens on127.0.0.1only (unless you open it withset hostname, and then only the extension link) and refuses cross-site browser requests and foreignHostheaders.Pairing. A pairing code works once, within 10 minutes; five wrong codes in a row void every pending one. A browser exchanges it for a random key of its own over a fresh key agreement (ECDH), so neither the code nor the key crosses the connection. On every later connection both sides prove they hold the key before the extension obeys anything, so a program that grabs the port first can't drive your browser. The relay accepts only the published extension IDs (plus any listed in
OPENCODE_BROWSER_EXTENSION_ORIGINS).executeis sandboxed, not a boundary. Scripts run in QuickJS inside the relay with CPU, memory and time limits, no Node, network or file access beyond the file roots (temp folders andOPENCODE_BROWSER_FILE_ROOTS). Treat code an agent runs as code running as you. The relay starts with an allowlisted environment, so your API keys aren't in it.CDP guardrails. Clearing cookies, cache or site data, closing the browser and changing site permissions are refused. Closing a page is handled like closing a target: a tab you lent is released, not closed. Cookie and storage access is limited to the hosts of the tab's own pages. Agents can't open browser-internal pages, other extensions or local files, or raise windows over your work.
Guardrails, not a data boundary.
allowedOriginslimits navigations, top-level documents and the tabs and popups an agent opens; requests a page makes itself (fetch, XHR, images, scripts, iframes) are not filtered.readOnlyrefuses clicks, hover, drag, typing, key presses, filling, selecting, uploads and navigation commands after the tab opens. Scrolling to an element or to the page's edges still works.evalandwebmcpstill run JavaScript in the page: they can read everything the page can, make requests as the page, and change it (click, submit, navigate). Snapshots, screenshots and the cookies the page itself can see stay readable. These keep an agent from wandering: for pages you don't want an agent to read, don't open them in browse.What output is masked.
fillhides values only for password and one-time-code fields. Snapshots leave out what's typed in editable fields, but not other page text. Network capture redacts credentials by name and shape (auth and API-key headers, cookie values, secret-named fields, token-like path segments), so a value under an ordinary name can still come through.evalresults, console messages, page text and screenshots are returned as they are.WebMCP tools come from the page. Their names, titles, descriptions, schemas, annotations,
originand results are whatever a page supplies; browse doesn't verify them, and an annotation likereadOnlyis only the page's word. Treat them like any other page text, not as instructions from the user.Bot challenges. After a navigation browse reports Cloudflare, captcha walls and "access denied" pages and tells the agent to hand interactive ones to you. It never solves them.
Your tabs stay yours. Agents use only tabs they opened or you lent them. Only a real click on the card or the toolbar button finishes a handoff, so a page can't fake it.
Escape hatches, set in the environment you start your agent from. The relay inherits OPENCODE_BROWSER_* variables
when it starts: it stops after 30 minutes without use, or run browse restart from a shell with the same variables:
Variable | Effect |
| Let tabs read and write every cookie in the profile |
| Let tabs use every site's storage, and load resources from any host with your credentials |
| Allow bringing a tab's window to the front |
| Allow |
| More folders tools may read and write (path-list separated) |
| Open agent tabs in a separate window ( |
| Send input at full speed; overrides the options page |
| Only tell about new versions ( |
| How long to wait for a bot check the browser clears by itself |
| How long the relay waits for the extension to answer a command |
| How long a ping may go unanswered before the browser counts as asleep |
| Pin the relay's port, with no fallback. The extension finds only browse's own ports (19988, 19990–19997) by itself, so enter |
| Which tools every client gets |
Troubleshooting
browse statusshows the install, the agents browse is set up in, updates, the relay, the extension and agent tabs, and says what to do when the extension is missing, unpaired, closed or out of date. Agents get the same from thestatustool.After the relay restarts, the extension reconnects by itself within about half a minute; calls made in the meantime wait for it.
An agent says browse's tools are missing: restart the agent (or reload its MCP servers) after
browse install."The browser isn't answering" while the computer is very busy: raise
OPENCODE_BROWSER_EXTENSION_TIMEOUT_MSandOPENCODE_BROWSER_PING_TIMEOUT_MS.Logs:
~/.local/state/browse/logs/relay.log.Another program's server on port 19988: browse takes the next free port from 19990–19997 and the extension finds it there. A port held by something that doesn't answer HTTP stops it; stop that program or set
OPENCODE_BROWSER_PORT.The extension says it's not paired, or the relay refused it: run
browse installagain, orbrowse pairand enter the code in the extension's options.An agent says browse was updated and its MCP server must restart: an MCP server started before an update to this version can't talk to the new relay. Restart the agent.
Development
bun install && bun run dev # rebuilds the extension into dist/ on change; load it unpacked, reload to pick it up
bun run typecheck && bun test ./test
cd rust && cargo test --workspace && cargo run -p browse -- statusA build from a checkout (cargo run -p browse -- install) is used in place and never updates itself. The extension in
dist/ keeps the manifest key, so it loads with the unpacked ID the relay trusts; run browse pair to pair it.
RELEASING.md has the release flow (dev builds from every push to dev; stable releases only from the
Release workflow in GitHub Actions), and rust/PLAN.md the layout of the Rust crates. cli/ is the
earlier TypeScript implementation, kept as a reference.
Contributing
We don't take pull requests. Open an issue describing the bug or idea, and our agents usually turn it around fast.
License
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Let AI agents use your real, signed-in Chrome: tabs, clicks, forms, screenshots and video.
Control a real browser from AI agents: navigate, click, type, screenshot, read the page.
Let your AI agent operate any website through a real, logged-in browser session.
Give AI agents a real cloud Chrome: open sessions, navigate, read pages, click, type and screenshot.
Related MCP Servers
- AlicenseBqualityDmaintenanceEnables AI agents to directly control your real Chrome browser with full context including login sessions, cookies, and open tabs. It provides tools for page scanning, JavaScript execution, CDP control, screenshots, and physical mouse/keyboard input for authentic browser automation.20243MIT
- AlicenseNot gradedqualityBmaintenanceEnables any coding agent to use the user's existing Chrome profiles and logins to open tabs, read pages, fill forms, click through flows, and take screenshots.21 npmMIT
- AlicenseNot gradedqualityAmaintenanceEnables AI assistants to drive the user's real, already-logged-in Chrome browser locally, so they can read and navigate pages, click, type, fill forms, take screenshots, and inspect cookies, tabs, console, and network traffic without headless browsers or remote-debugging ports.1 npmISC
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to drive your real browser — your tabs, logins, and extensions — by opening and controlling pages with Playwright, taking screenshots, capturing network traffic and recordings, and handing tabs back to you for logins or payments. Works across Chrome, Edge, Brave, Arc, Helium, Vivaldi, Opera, and Chromium through a local relay and extension.16,934 npm4MIT