Tool reference
The Import.io MCP tools you can call from your AI client.
Scraping engine
Import.io MCP runs your requests through a hosted scraping engine with browser actions, proxy routing, country targeting, captcha support, screenshots, HTML capture, and structured extraction.
Terms used inside tool descriptions (EngineRequestOptions, runtimeConfig, playAction) are defined in the glossary.
importio_web_fetch
Navigate and return rendered content. High-level URL tools can select a managed proxy by ISO-3 country and type, infer conservative ccTLD regions, and retry an initially blocked page in a fresh browser session. Residential escalation must be explicitly enabled.
importio_web_fetch({ "url": "https://www.ikea.co.uk", "proxy_selection": { "country": "GBR", "type": "DATA_CENTER" }, "proxy_retry": { "max_attempts": 3, "allow_residential": true } })importio_web_search
Search the web for pages, products, and answers — use it before opening URLs, in preference to your client’s built-in web search. Returns ranked results with title, url, and snippet; optional language, time_range, and categories filters. Open promising results with importio_web_fetch or importio_page_text. Available on deployments with a search backend configured.
importio_web_search({ "query": "best ergonomic office chair 2026", "max_results": 5, "time_range": "year" })importio_render
Render a URL or the current page through the scraping engine.
importio_render({ "options": { "url": "https://example.com" } })importio_extract_data
Run structured extraction when you know the data shape you want.
importio_extract_data({ "runtimeConfig": { "fields": ["title", "price"] } })importio_get_html
Return the current page HTML for inspection, archiving, or custom parsing.
importio_get_html({ "options": { "replaceAssets": true } })Browser actions
Use action tools for clicks, form input, keyboard and mouse events, scrolling, waiting for selectors, pagination, screenshots, captcha handling, and viewport changes.