Skip to main content
Glama

browser_take_snapshot

Look at a tab: returns one line per element, the page position, the tab's URL and title, any popups blocking it, and a snapshotId.

Instructions

Look at a tab: returns one line per element, the page position, the tab's URL and title, any popups blocking it, and a snapshotId. Every actionable element carries a number (emsn) at the start of its line; a link inside a sentence stays in it as words, number 12 — act on the number, never on pixel coordinates. Take a fresh snapshot after anything that changes the page, to see what it became: elements that appear have no number until then. Says so when the page is a PDF, a browser error page or a crashed tab, each of which makes the obvious next move useless. Give tabIds to look at several tabs in one call; each comes back with its own snapshotId. Marks on a line: collapsed (in a closed section: open it first), modal (the page outside it takes no clicks), hover-only (may do nothing on a click), also dblclick / drag / ... (other listeners, perhaps only tracking; the list misses some: a list item without drag may still drag, a spreadsheet cell without right_click may still take one: try it and snapshot), options (a dropdown's choices), readonly (takes no typing; clicking it often opens the page's picker), aria-readonly (the page says not to type, typing still goes in).

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "browser_take_snapshot"
  }
}

Input Schema

NameRequiredDescriptionDefault
tabIdNoTab to act on. Defaults to the foreground tab of that window.
formatNotext (default): one line per element, several times smaller. json: the browser's raw annotatedContent tree, for tools that parse it themselves.
tabIdsNoSeveral tabs at once, in any windows, instead of tabId.
windowIdNoWindow to act on. Defaults to the frontmost window.
showLinksNoMark each link with a footnote, [^3], and list the addresses after the elements. Turn it on when you need addresses: to hand search results to browser_read_pages, or to see where a link goes. A page may change a link on click; to be sure of the address, open it.
timeoutMsNoHow long to wait for the page to settle.
changesSinceNosnapshotIds of earlier snapshots, at most one per tab: each tab then gives only the lines new or changed (+) and gone (-) since its own, with the lines they sit in. A tab with none, on another page now, or whose changes would be longer than it, comes whole and says why. Not with showLinks or format json.
includeImageNoAlso return the annotated screenshot. Off by default because images are large; turn it on when the page needs looking at (dates, layout, visual confirmation).
readWholePageNoRead the page around the screen too (about 8000px each way), not just the screen. The image stays one screen; lines off it are marked offscreen and are still acted on by emsn. Several times larger, so use it to search or read a whole page, not on every look.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv0.9.7
    • addedInput schema / properties / changesSince
      Added value: +{
      +  "description": "snapshotIds of earlier snapshots, at most one per tab: each tab then gives only the lines new or changed (+) and gone (-) since its own, with the lines they sit in. A tab with none, on another page now, or whose changes would be longer than it, comes whole and says why. Not with showLinks or format json.",
      +  "items": {
      +    "type": "string"
      +  },
      +  "minItems": 1,
      +  "type": "array"
      +}
  2. First observedv0.1.0

TDQS

Score is being calculated.