Skip to content

Latest commit

 

History

History
18 lines (15 loc) · 2.43 KB

browser.md

File metadata and controls

18 lines (15 loc) · 2.43 KB

Browser & API Discovery

No pre-built integration needed — these work on any site you're logged into.

Tools

Tool Description Parameters
extract_auth Extract auth tokens from the user's logged-in browser session. Supports: slack, discord, linkedin, twitter, github, notion, or any domain. Tokens are stored automatically for future API calls. Use the profile parameter to store credentials under a named profile (e.g. profile='business' stores as 'linkedin:business'). service
authenticated_fetch Make an HTTP request from the browser's context, carrying the page's cookies, auth, and session. Works on ANY website the user is logged into. This is the meta-tool for building integrations on the fly. If no pre-built tool exists for a service: 1. Use discover_api to find the site's API endpoints 2. Use authenticated_fetch to call them 3. Use collection_create to save the discovered API pattern (endpoint, method, headers) so you can reuse it next time without rediscovering url, method?, headers?, body?
network_capture Start/stop/clear network request capture in the browser. Use network_requests to list and network_request_detail to inspect. action, filters?, navigate?
network_requests List captured network requests. Returns id, method, status, URL. Use network_request_detail to get full headers/body for a specific request. filter?, limit?
network_request_detail Get full details for a captured request — request headers, response headers, and body. Pass the id from network_requests. id
bridge_status Check if the Neo Browser Bridge extension is connected. —
web_scrape Extract structured data from any URL. Returns clean, parsed content instead of raw HTML. Extracts: page title, meta description, main text content, all links, tables (as arrays), images, OpenGraph/meta tags, and JSON-LD structured data. Use this instead of authenticated_fetch when you need usable data from a page. url, extract?, selector?
diff_monitor Monitor any URL or API endpoint for changes. Stores snapshots in a collection, compares against the previous snapshot, and reports what changed. Use for: price tracking, job posting changes, competitor monitoring, stock availability, or any "tell me when X changes" request. action, url?, name?, selector?, extract?