lineage
Extract only passages from documents or web pages that answer your questions, resolving people, places, organisations and dates in context.
Instructions
Read a document or a web page and return only the passages that answer your questions, with the people, places, organisations and dates written in where a passage needs them (the name behind 'she', an exact date for 'last April'). Send text or a url, and the questions in queries. Each question is a separate pass and is billed as one. Without queries, every passage comes back, which is larger than the input. Use it to turn a long source into grounded context.
Tool call template
This tool declares no required parameters. Add optional parameters from the input schema as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {},
"name": "lineage"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| url | No | A page for Flux to fetch and read instead of text. If the site refuses, you get a bad_request error and are not charged. | |
| text | No | The document to read, up to 60000 characters. Use this or url. | |
| query | No | One question. Same as queries with a single entry. | |
| queries | No | The questions the passages should answer, up to 10. Each is a full pass and billed as one, so ask only what you need. Omit to get the whole document. | |
| documents | No | Several documents in one call, each with its own text or url. Use instead of text/url. Top-level queries apply to every document unless a document has its own. Documents that cannot be read are listed and not charged. | |
| published | No | When the text was published, as an ISO date (2026-03-14). Lets relative dates such as 'last April' resolve to the right year. Not needed for a url that states its date. |