Skip to content
Agentmart
Submit a tool

Browser Gateway

AI browser tools with Chrome CDP.

Connector by browser-gateway · GitHub account verified by the MCP Registry

IndexedMITLocal (stdio)★ 18
Write a review

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/browser-gateway-browser-gateway

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

AI browser tools with Chrome CDP. Navigate, screenshot, interact. Multi-provider failover.

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list, install config and code. Checked Oct 25, 2026.

Tools
7 read · 3 write · 0 delete · 3 not yet classified
Secrets
1 (API keys, tokens or passwords)
Install scripts
None found
Runs as
Node.js package

Tools by what they do 13

Read from the source code; the running server may list different tools.

Read 7 · Look things up without changing anything.

  • browser_extract
  • browser_screenshot
  • browser_snapshot
  • browser_status
  • browser_tabs
  • fetch_page
  • screenshot_page

Write 3 · Create, change, send or run things.

  • browser_act
  • browser_navigate
  • browser_session

Not yet classified 3 · We could not tell from the name and description.

  • browser_evaluate
  • browser_observe
  • browser_wait

Secrets and settings it asks for

Secrets: values you paste in that let it act as you.

  • BG_TOKEN

Network hosts it talks to 13

  • browsergateway.com · found in code
  • cdp.browsergateway.io · found in code
  • developer.microsoft.com · found in code
  • docs.browsergateway.com · found in code
  • github.com · found in code
  • glama.ai · found in code
  • google.com · found in code
  • iana.org · found in code
  • ipinfo.io · found in code
  • rest.invalid · found in code
  • ui.shadcn.com · found in code
  • unpkg.com · found in code
  • your-app.com · found in code

Files it touches

  • %APPDATA%\\devin\\mcp_config.json\
  • %USERPROFILE%\
  • %USERPROFILE%\\.codex\\config.toml\
  • Your home directory (~)

Tools (13)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

  • browser_act

    Run one or more steps (click, fill, type, press, hover, check, uncheck, select, scroll, wait) and return only what changed.

  • browser_evaluate

    Run JavaScript in the page and return its value. Save a long script once with saveAs, then run it by name with script on later pages instead of sending it again. Saved scripts last as long as this browser.

  • browser_extract

    Read the current page in the open browser as markdown, plain text or a link list. Much cheaper than a screenshot. To read a url without opening a session, use fetch_page.

  • browser_navigate

    Go to a url in the open browser session and list the page's clickable and typeable elements, labelled e1, e2. To run your own script on each page, pass it in evaluate (or a saved one by script) and set snapshot false to get only its result.

  • browser_observe

    Console output, failed requests, downloads, dialogs and popup tabs seen in this session.

  • browser_screenshot

    Screenshot the current page in the open browser, or one element by its label. To screenshot a url without opening a session, use screenshot_page.

  • browser_session

    Start a real browser for multi-step work on a website: logging in, filling forms, clicking through pages, reading several pages. For a single page read or screenshot, use fetch_page or screenshot_page instead. One browser per task; it closes after a few idle minutes, so close it yourself when done.

  • browser_snapshot

    List what you can click or type on the page, labelled e1, e2. Labels change after every navigation.

  • browser_status

    Gateway providers, sessions and queue state.

  • browser_tabs

    List, open, switch or close tabs in this session.

  • browser_wait

    Wait for text or an element to appear or disappear, or for the url to change.

  • fetch_page

    Read a web page with a real browser and get clean markdown, in one call. Use when a built-in web fetch failed (403, 402, 429, captcha, empty or JavaScript-only page) or the page needs JavaScript or a saved login. For ordinary static pages, try the built-in web fetch first. Opens a browser, reads the page and closes it. Returns `blocked` for a bot check, error or login wall. JSON and plain text come back raw, with `contentType`.

  • screenshot_page

    Take a screenshot of a website or web page with a real browser, in one call. Give a url, get the image. Use for "take a screenshot of", "what does this site look like" or checking a layout, including JavaScript-heavy pages. Opens a browser, loads the page, captures it and closes it. For several steps on one site, use browser_session instead.

Signals

Stars
18
Forks
5
Open issues
0
Last commit
Oct 20, 2026 (8 days ago)
Last release
Oct 20, 2026 (8 days ago)

From GitHub (fetched Oct 24, 2026).

Details

Registry name
io.github.browser-gateway/browser-gateway
Version
0.4.34
Auth
API key
Packages
npm browser-gateway
License
MIT
Repository
github.com/browser-gateway/browser-gateway
First seen
Oct 24, 2026
Last checked
Oct 25, 2026
Source
Official MCP Registrygithub-graphql
Cite this page

Browser Gateway MCP server. Agent Mart. https://agentmart.dev/connectors/browser-gateway-browser-gateway (last updated 2026-10-25)