Linkup Fetch
When you already have the exact URL, use Fetch instead of search. It is faster and cheaper, and returns the page as clean Markdown (with JavaScript rendering and optional image extraction).
How to call it
- If the
linkup-fetchMCP tool is available, pass it the URL. - Otherwise call the REST API directly. Requires
LINKUP_API_KEY:
curl -sS -X POST "https://api.linkup.so/v1/fetch" \
-H "Authorization: Bearer $LINKUP_API_KEY" -H "Content-Type: application/json" \
-d '{"url":"https://example.com/pricing","renderJs":true}'
Default JavaScript rendering to on — many sites load content client-side, and the small latency cost is almost always worth the reliability. Turn it off only when speed matters more on a known-static page.
When to use fetch vs the alternatives
Use linkup-fetch when... |
Use instead... |
|---|---|
| You have one URL and want its content as Markdown | — |
| You don't know which URL has the answer | linkup-search |
| You need many structured records from one listing page (team, catalog, jobs) | linkup-extract (/v1/extract) |
| You need to discover URLs and then scrape them | linkup-search with depth: deep |
| The URL is a LinkedIn profile or post | linkup-search (Fetch cannot authenticate into LinkedIn) |
Important: fetch reads page content; it cannot produce structured JSON from a page. If you need structured fields from a known page, use linkup-extract (/v1/extract) or the Search API with outputType: structured.
After fetching
- Extract only the fields the task needs; don't dump the whole page back to the user.
- Keep the source URL so every claim can be verified.
For the full endpoint reference, read references/LINKUP_API_REFERENCE.md (Fetch API section) in this skill's directory.