Page

Page — MCP tools exposed to agents.

20 tools in this group.

browser_open

Navigate the browser to a URL. url is required; a bare host (e.g. example.com) is prefixed with https://. To launch (or connect to) a browser without navigating, use browser_start; to close it, use browser_stop. (The CLI open is url-optional and lazy-launches when bare — this agent tool is url-required by design: same capabilities, different surface ergonomics, not a 1:1 syntax mirror.)

ParameterTypeRequiredDefaultDescription
sessionstringnoOptional: named session to run this call in (default: current session).
urlstringyesThe URL to navigate to (required; bare hosts get https:// prepended)

browser_back

Navigate back in browser history (like clicking the back button)

ParameterTypeRequiredDefaultDescription
sessionstringnoOptional: named session to run this call in (default: current session).

browser_forward

Navigate forward in browser history (like clicking the forward button)

ParameterTypeRequiredDefaultDescription
sessionstringnoOptional: named session to run this call in (default: current session).

browser_reload

Reload the current page. Waits for the page to fully load.

ParameterTypeRequiredDefaultDescription
ignore_cachebooleannoBypass the browser cache when reloading (default: false)
sessionstringnoOptional: named session to run this call in (default: current session).
waitstringnoReadiness state to wait for before returning (default: complete) (one of: none, interactive, complete)

browser_get_url

Get the current page URL

ParameterTypeRequiredDefaultDescription
sessionstringnoOptional: named session to run this call in (default: current session).

browser_get_title

Get the current page title

ParameterTypeRequiredDefaultDescription
sessionstringnoOptional: named session to run this call in (default: current session).

browser_get_text

Get the text content of the page or a specific element

ParameterTypeRequiredDefaultDescription
selectorstringnoCSS selector for a specific element (optional, defaults to full page text)
sessionstringnoOptional: named session to run this call in (default: current session).

browser_get_html

Get the HTML content of the page or a specific element

ParameterTypeRequiredDefaultDescription
outerbooleannofalseReturn outerHTML instead of innerHTML (default: false)
selectorstringnoCSS selector for a specific element (optional, defaults to full page HTML)
sessionstringnoOptional: named session to run this call in (default: current session).

browser_read

Return a page as agent-readable text (markdown). With a url, fetches it over HTTP without the browser (negotiates markdown/plain, retries a .md path, discovers llms.txt, extracts readable text from HTML). WITHOUT a url, renders the active tab's live DOM, capturing client-rendered/authenticated state a plain fetch cannot.

ParameterTypeRequiredDefaultDescription
filterstringnoNarrow output to sections/headings/links matching this text
jsonbooleannofalseReturn the full fetch envelope (url, finalUrl, status, contentType, source, content) as JSON instead of just the content
llmsstringnoRead the nearest llms.txt (index) or llms-full.txt (full) instead of the page (one of: index, full)
outlinebooleannofalseReturn a compact heading outline of the page
rawbooleannofalseReturn the body verbatim, without markdown extraction
require_mdbooleannofalseFail unless the response Content-Type is text/markdown
sessionstringnoOptional: named session to run this call in (default: current session).
timeout_msintegernoRequest timeout in milliseconds (default 30000)
urlstringnoURL to fetch (http/https; a bare host assumes https). Omit to read the active tab's live DOM instead.

browser_set_content

Replace the page HTML content

ParameterTypeRequiredDefaultDescription
htmlstringyesHTML content to set
sessionstringnoOptional: named session to run this call in (default: current session).

browser_a11y_tree

Get the accessibility tree of the current page. Returns a tree of ARIA roles, names, and states — useful for understanding page structure without visual rendering.

ParameterTypeRequiredDefaultDescription
everythingbooleannofalseShow all nodes including generic containers. Default: false
sessionstringnoOptional: named session to run this call in (default: current session).

browser_evaluate

Execute JavaScript in the browser to extract data, query the DOM, or inspect page state. Returns the evaluated result. Use this to get text content, attributes, element data, or any information from the page.

ParameterTypeRequiredDefaultDescription
expressionstringyesJavaScript expression to evaluate
sessionstringnoOptional: named session to run this call in (default: current session).

browser_wait

Wait for an element to reach a specified state (attached, visible, or hidden)

ParameterTypeRequiredDefaultDescription
selectorstringyesCSS selector for the element to wait for
sessionstringnoOptional: named session to run this call in (default: current session).
statestringno"attached"State to wait for: "attached" (exists in DOM), "visible" (visible on page), or "hidden" (not found or not visible) (one of: attached, visible, hidden)
timeoutintegerno30000Timeout in milliseconds (default: 30000)

browser_wait_for_url

Wait until the page URL contains a given substring

ParameterTypeRequiredDefaultDescription
patternstringyesSubstring to match in the URL
sessionstringnoOptional: named session to run this call in (default: current session).
timeoutintegerno30000Timeout in milliseconds (default: 30000)

browser_wait_for_load

Wait until the page reaches the "complete" ready state (all resources loaded)

ParameterTypeRequiredDefaultDescription
sessionstringnoOptional: named session to run this call in (default: current session).
timeoutintegerno30000Timeout in milliseconds (default: 30000)

browser_wait_for_text

Wait until specific text appears on the page

ParameterTypeRequiredDefaultDescription
sessionstringnoOptional: named session to run this call in (default: current session).
textstringyesText to wait for on the page
timeoutintegerno30000Timeout in milliseconds (default: 30000)

browser_wait_for_fn

Wait until a JavaScript expression returns a truthy value

ParameterTypeRequiredDefaultDescription
expressionstringyesJavaScript expression to evaluate (e.g., "window.ready === true")
sessionstringnoOptional: named session to run this call in (default: current session).
timeoutintegerno30000Timeout in milliseconds (default: 30000)

browser_console_list

List buffered browser console messages and page errors (WebDriver BiDi log.entryAdded). Optional filters narrow the result without clearing the buffer.

ParameterTypeRequiredDefaultDescription
levelstringnoOnly include entries at this log level (one of: debug, info, warn, error)
sessionstringnoOptional: named session to run this call in (default: current session).
textstringnoOnly include entries whose text contains this substring
typestringnoOnly include console API entries or javascript exception entries (one of: console, javascript)

browser_console_wait

Wait for the next matching browser console message or page error (log.entryAdded). Returns immediately if a matching buffered entry already exists; consumes the matched entry.

ParameterTypeRequiredDefaultDescription
levelstringnoOnly match entries at this log level (one of: debug, info, warn, error)
sessionstringnoOptional: named session to run this call in (default: current session).
textstringnoOnly match entries whose text contains this substring
timeout_msintegerno30000Timeout in milliseconds (default: 30000)
typestringnoOnly match console API entries or javascript exception entries (one of: console, javascript)

browser_console_clear

Clear the buffered browser console / page-error history for this session

ParameterTypeRequiredDefaultDescription
sessionstringnoOptional: named session to run this call in (default: current session).

On this page