live=true and fetch reads the page from its source on this call. Use it for pages that change between visits: status pages, price lists, dashboards, schedules.
Parameters
Fetch a status page live
This sample reads the GitHub status page from its source and uses aprompt to return only the components that aren’t operational. It runs with no API key:
Example output · results vary
Response · results vary
Indexed copy or live read
Leave
live out for most pages. Documentation, articles and papers rarely change between the time Keenable indexed them and your call. A search result’s acquired_at says when Keenable indexed that page, so you can compare it with how often the page changes.
See which one you paid for
With an API key, a successful response carriesusage. Its sku is fetch for the indexed copy and fetch.live for a live read, whether you set live=true or Keenable had no indexed copy. The keyless endpoint leaves usage out. See Balance and usage and Pricing.
Common mistakes
In the SDKs, CLI and MCP
fetch takes live in every SDK: live=True in Python, fetch(url, { live: true }) in TypeScript (from keenable 0.2.0), FetchOptions.Live in Go and FetchOptions.live in Rust. See SDK methods. The CLI takes --live (see CLI). Over MCP, fetch_page_content takes live (see MCP server).
API reference
- Fetch request:
liveand the other fetch parameters. - Fetch: what each fetch costs.
- Extract part of a page with a prompt: write the prompt for a live read.