Michael Hablich comments

Results 42 comments of


                                            Michael Hablich

User agent modification support

Relevant: https://github.com/ChromeDevTools/chrome-devtools-mcp/issues/395

User agent modification support

This is also needed for proper device emulation.

FR: Could we add BODY_CONTEXT_SIZE_LIMIT to the env, or add a param to the get_network_request?

This should likely work similar to how we send back screenshots. If they are over a limit, we store them in a file.

Allow changing default arguments

https://github.com/ChromeDevTools/chrome-devtools-mcp/pull/338 should have solved this, right?

Allow changing default arguments

@OrKoN does the --no prefix work for this use case?

return all pages when click web element

I think this needs to be optional, not every click results in a reload of the whole tree and returning the whole tree will just consume tokens like crazy.

Is it possible to call the zoom in/out function of chrome browser

Have you tried letting the agent use evaluate_script with ```document.body.style.zoom = 0.5;```?

Allow Toggle Device Toolbar

viewport, cpu and network throttling can already be set via emulate and resize_page. Have you tried simply saying in the prompt something like "Open this URL example.com, emulate that your...

A tool to "Select an element in the page to inspect it"

Do you mean that you can select an element/artifact in Chrome DevTools and use it as context in your MCP client?

A tool to "Select an element in the page to inspect it"

Understood, so the human would then use the selected element to continue debugging on their own in DevTools?