{"version":"2.1.290","anchor":"webfetch-gains-an-offset-parameter-for-paging-long-pages","canonical_anchor":"webfetch-gains-an-offset-parameter-for-paging-long-pages","heading":"WebFetch can continue reading a long page from an offset","tier":"notice","area":"Web Fetch","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/webfetch-gains-an-offset-parameter-for-paging-long-pages","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290","markdown":"### WebFetch can continue reading a long page from an offset\n\nWebFetch gains an optional `offset` so Claude can read on through a page too long for one fetch instead of only seeing its start\n\n**Unclear.** It is not clear whether the tool already uses the offset when fetching, or only accepts it for now.\n\n**What**\n\nWebFetch is the tool Claude uses to read a web page. It used to take only a `url` and a `prompt`, so on a very long page Claude always got the same cut-off first part. It now accepts an optional `offset`:\n\n- `offset` is a character position in the page text, a whole number of zero or more. Claude passes the value the previous result gave to pick up where it left off.\n\n- A page shown from a given position carries a \"from character N\" note.\n\n- Fetching again at the same offset returns the same split.\n\n- If the offset is past the end of the page, the result says nothing is left to show from that offset.\n\n- When Claude Code shows tool calls in detail, the WebFetch line includes the offset when one was used.\n\nIn the prompt capture taken under this release, the tool's stock description is unchanged from the previous version.\n\n**Why**\n\nClaude can now read a long page in pieces instead of stopping where the first piece was cut off. If you watch the tool calls, an `offset:` value means Claude is reading further into the same page.\n\n- Area: Web Fetch\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}