browser_snapshotLABSCO SUMMARY
Browser automation with the debugging reads built in.
Beyond clicking and typing, it reads what a QA pass needs: a diff between two snapshots, console errors with stack traces, the page's forms, storage keys and cookies, and full-page screenshots. The extension talks only to a localhost server, so browsing data stays on your machine.
What it is
A browser-driving server built around accessibility snapshots, paired with a Chrome extension, with tools for navigation, input, waiting and readback.
What you get
- Move around: open a URL, go back, go forward.
- Take an accessibility snapshot of the page, and compare the two most recent snapshots to see what changed.
- Act on an element picked by CSS selector — click, hover, drag it onto a target, type with an optional Enter, choose dropdown options, or fill several inputs in one call.
- Press a key, pause for a set number of seconds, or wait until a selector appears.
- Read the result back as console output with stack traces, a viewport or full-page screenshot, or a dump of the page's forms, storage keys and cookies — and run your own JavaScript or answer dialogs when those are not enough.
Requirements
Chrome with the YetiBrowser extension installed and connected to a tab, and npx to run @yetidevworks/server. No account, no key.
Setup effort
One command — npx -y @yetidevworks/server
USE CASES
01
Filling in a form on a page you cannot script against
Open the page, snapshot it to get references, then type into the fields and pick the dropdown values before submitting.
02
Reproducing a bug report that only happens in the browser
Walk the reported path, pull the console logs for the errors the reporter never mentioned, and capture the screen at the point it breaks.
03
Walking a multi-step checkout to find where it stalls
Click through each step, give the page time to settle, re-snapshot for the new references, and back up a step when the flow dead-ends.
04
Checking that a hover-only menu really opens
Put the pointer on the trigger, snapshot to see whether the menu entered the accessibility tree, and keep a picture of what appeared.
20 tools
Listed in project docs. Your client may show a different list.
browser_snapshot_diffbrowser_navigatebrowser_go_backbrowser_go_forwardbrowser_waitbrowser_wait_forbrowser_press_keybrowser_clickbrowser_hoverbrowser_dragbrowser_typebrowser_fill_formbrowser_select_optionbrowser_screenshotbrowser_get_console_logsbrowser_page_statebrowser_connection_infobrowser_evaluatebrowser_handle_dialogBEFORE YOU INSTALL
What to know first
Tools you get
20 tools named in the project's documentation.
Requires
npx on PATH
INSTALLATION
Connect your client
Server setup
TERMINAL
npx -y @yetidevworks/serverMCP Clients
Codex
CODEX · TOML
[mcp_servers.yetibrowser-mcp]
command = "npx"
args = ["-y", "@yetidevworks/server"]Claude Code
CLAUDE CODE · TERMINAL
claude mcp add yetibrowser-mcp -- npx -y @yetidevworks/serverMCP Inspector
MCP INSPECTOR · TERMINAL
npx @modelcontextprotocol/inspector npx -y @yetidevworks/server -- --ws-port 9010
