Create a link context item by extracting / crawling a URL
Invokes the Tavily access Lambda to extract content from the supplied URL (single-page) or crawl up to pageLimit pages when isCrawl is true. The extracted markdown is stored in tenant S3; only metadata is kept on the context_items row.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Body
URL to extract content from. Invoked via Tavily.
"https://example.com/blog/post"
Optional title override. When omitted, Tavily-derived metadata or the URL hostname is used.
"Example blog post"
When true, crawls multiple pages starting from the supplied URL.
false
Maximum number of pages to crawl when isCrawl is true. Ignored otherwise.
1 <= x <= 20010
Response
UUID of the newly-created context item. Use it on subsequent GET / PATCH / DELETE calls.
"7a3f9d2c-1111-4000-a000-000000000abc"
Kind of context item created.
note, saved_query, link, dimension, fact, metric, filter "note"
Tenant S3 key for the raw content blob (notes, saved queries, links). Null for kinds without S3 backing.
"context_notes/7a3f9d2c.../Q3_churn_analysis.md"
Resolved title for the created item. Null for kinds without a title field.
"Q3 churn analysis"