browser_click
Click an element on the page, before using the tool, use browser_get_clickable_elements to get the index of the element, but not call browser_get_clickable_elements multiple times
LABSCO SUMMARY
Structured page data instead of pixels — which is why it works on models with no vision at all.
Elements are addressed by index from a single clickable-elements listing, so a page costs a fraction of what a screenshot loop costs in context. When structure genuinely is not enough — a canvas, a chart, a visual layout — --vision switches it over to screenshots for that run. browser_evaluate is the escape hatch for everything the tool list does not name.
What it is
A Puppeteer-based browser automation server that works from the page's structured accessibility data and element indexes rather than pixels, with an optional screenshot mode for layouts where structure is not enough.
What you get
- Navigation: open a URL, move back and forward, and shut the browser down when the task ends — browser_navigate, browser_go_back, browser_go_forward, browser_close
- Interaction by element index, after listing the page's clickable elements once — browser_get_clickable_elements, browser_click, browser_hover, browser_select, browser_form_input_fill
- Keyboard and scrolling — browser_press_key, browser_scroll
- Page content back as markdown or plain text, and every link on the page — browser_get_markdown, browser_get_text, browser_read_links
- Screenshots of the full page or one element, with an option to highlight it — browser_screenshot
- Tabs opened, listed, switched and closed — browser_new_tab, browser_tab_list, browser_switch_tab, browser_close_tab
- JavaScript executed in the page console — browser_evaluate
- The files downloaded during the session — browser_get_download_list
Requirements
No account and no key. Node.js 18 or newer, and npx to run @agent-infra/mcp-server-browser over stdio; adding --port turns the same binary into a remote server that serves Streamable HTTP at /mcp and SSE at /sse. --browser chooses chrome, edge or firefox, --executable-path points at a specific binary, and --cdp-endpoint or --ws-endpoint attaches to a browser that is already running. --proxy-server and --proxy-bypass route its traffic, --vision switches it from the default Aria snapshots to screenshots, and DISPLAY covers rendering on a virtual display.
Setup effort
One command — npx @agent-infra/mcp-server-browser@latest
USE CASES
01
Reading a page that only exists after JavaScript has run
Open the URL in the real browser and pull the rendered page back as markdown.
02
Filling in a form and submitting it
List the clickable elements once, fill the fields by index, then click the button.
03
Walking a multi-step flow to see where it breaks
Click through the steps and go back when a branch turns out to be a dead end.
04
Checking where a page's links actually lead
Read every link, open the interesting ones in their own tabs, and switch between them.
05
Capturing what the page looked like at the time
Screenshot the whole page, or just the element in question with the highlight on.
06
Getting at something no tool covers
Run the JavaScript in the page console and read the result back as text.
21 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
browser_close
Close the browser when the task is done and the browser is not needed anymore
browser_close_tab
Close the current tab
browser_evaluate
Execute JavaScript in the browser console
browser_form_input_fill
Fill out an input field, before using the tool, Either 'index' or 'selector' must be provided
browser_get_clickable_elements
Get the clickable or hoverable or selectable elements on the current page, don't call this tool multiple times
browser_get_download_list
Get the list of downloaded files
browser_get_markdown
Get the markdown content of the current page
browser_get_text
Get the text content of the current page
browser_go_back
Go back to the previous page
browser_go_forward
Go forward to the next page
browser_hover
Hover an element on the page, Either 'index' or 'selector' must be provided
browser_navigate
Navigate to a URL
browser_new_tab
Open a new tab
browser_press_key
Press a key on the keyboard
browser_read_links
Get all links on the current page
browser_screenshot
Take a screenshot of the current page or a specific element
browser_scroll
Scroll the page
browser_select
Select an element on the page with index, Either 'index' or 'selector' must be provided
browser_switch_tab
Switch to a specific tab
browser_tab_list
Get the list of tabs
BEFORE YOU INSTALL
What to know first
Tools you get
21 tools when we connected; the project documents 23. The two extra tools, browser_vision_screen_capture and browser_vision_screen_click, appear only when the server starts with --vision; by default it uses Aria snapshots instead of screenshots.
Requires
- Node.js 18 or newer
- VS Code, Cursor, Windsurf, Claude Desktop or any other MCP client
INSTALLATION
Connect your client
Server setup
TERMINAL
npx @agent-infra/mcp-server-browser@latestMCP Clients
VS Code
VS CODE · TERMINAL
code --add-mcp '{"name":"browser","command":"npx","args":["@agent-infra/mcp-server-browser@latest"]}'Cursor, Windsurf, Claude Desktop
JSON
{
"mcpServers": {
"browser": {
"command": "npx",
"args": [
"@agent-infra/mcp-server-browser@latest"
]
}
}
}
