Tool Center
Chromium Browser
Seeing a page the way a person would: rendered, after its JavaScript has run, with what the console said -- and a picture of it the Resident can actually look at. Drives Chromium, from this capability's own folder or from this machine.
Tools
close_browser close_browser.pyClose the browser and free what it is holding. The next open_page starts a fresh one.
install_engine install_engine.pyDownload the Chromium this capability needs, if the machine has none. Fetches Chrome for Testing's headless shell (about 100 MB) into this capability's own folder. Call this when open_page says no Chromium was found.
open_page open_page.pyOpen a URL in a real browser and read the page after its JavaScript has run. Returns the page's text, its links and controls, and anything the console or the network complained about. The page stays open for read_page, page_do and screenshot.
page_do page_do.pyAct on the page that is open: click something, type into a field, press a key, scroll, or wait for something to appear. Takes a list of steps, does them in order, then reads the page back. Elements are found by ref, CSS selector or the text on them.
page_requests page_requests.pyList what the open page fetched and what came back: method, URL, status, type and size. 'contains' narrows it to one endpoint, 'body' reads that response.
read_page read_page.pyRead the page that is currently open, without loading it again. Give a CSS selector to read one part of it, or nothing for the whole page.
screenshot screenshot.pyTake a picture of the page that is open. It is saved as a PNG in the Living Workspace and you are shown it. for_owner puts it in the chat for the owner; otherwise it is folded away with the other results.
No description beyond the summary above.