Zyte api MCP integration

It runs an HTTP fetch, full browser rendering, or AI-powered extraction against a URL, and can return structured records such as products, articles, and job posts, a screenshot of the page, or a custom schema you define. It also supports geolocation targeting, so a page can be fetched the way a visitor in a given country sees it.

Because that is what it exposes: one extraction action and six reads of Zyte's status API. There is no action for managing spiders, jobs, or Scrapy Cloud runs, so scheduling and crawl management stay in Zyte's own console.

Zyte API Components returns each component with a status of operational, degraded_performance, partial_outage, or major_outage, plus its description and grouping. That is enough for Lyro to say whether the specific part a workflow depends on is the affected one.

Lyro reads the status rollup and the unresolved incident list, says what is affected, and hands the conversation to a human agent with the full history when the request cannot be completed. It runs on top of your existing help desk, so the thread stays where your team already works.