Connection check
local install
mcp-textbrowser
MCP server — text-first headless browser (DOM + OCR). Zero image tokens by default; 5-15x cheaper than screenshot-based browser MCPs.
Tools
7
GitHub stars
1
Installs / wk
25
Licence
MIT
Transport
stdio
Last checked
never
Tools & capabilities
7 toolsRead from the published package source — this server runs locally, so there is no endpoint to query. Names are taken from the code, not observed at runtime, and descriptions are often absent.
browser_click
from source
Click an element on the current page. Identify the target by CSS selector, XPath, or visible text. Returns updated page context after the click. Click an element on the current page. Identify the target by CSS selector, XPath, or visible text. Returns updated page context after the click.
browser_evaluate
from source
Execute JavaScript in the current page context and return the result. Restricted to safe DOM read operations \u2014 eval, Function, fetch, WebSocket, and document.write are blocked… Execute JavaScript in the current page context and return the result. Restricted to safe DOM read operations \u2014 eval, Function, fetch, WebSocket, and document.write are blocked.
browser_navigate
from source
Open a URL in the browser and return the page context (DOM elements + OCR text). Default text-only mode \u2014 screenshot is captured only for OCR then discarded, zero image tokens… Open a URL in the browser and return the page context (DOM elements + OCR text). Default text-only mode \u2014 screenshot is captured only for OCR then discarded, zero image tokens. Set visual=true to also receive the PNG.
browser_read
from source
Read the current page context (DOM elements + OCR text) without navigating or clicking anything. Read the current page context (DOM elements + OCR text) without navigating or clicking anything.
browser_screenshot
from source
Capture the current page and return an OCR-based text map. The screenshot itself is discarded unless visual=true. Capture the current page and return an OCR-based text map. The screenshot itself is discarded unless visual=true.
browser_scroll
from source
Scroll the current page. Use direction (up/down/left/right) with an optional pixel amount, or provide a selector to scroll that element into view. Scroll the current page. Use direction (up/down/left/right) with an optional pixel amount, or provide a selector to scroll that element into view.
browser_type
from source