Liner icon

Liner

Search the web and generate source-backed answers with Liner

Actions7

Answer → Deep Research

AI-generated

Summary

Generate a long-form research report based on a user question or a conversation history using Liner's Deep Research API.

Inputs

  • Language — Optional language code (e.g., en, ko, or ja) to specify the language for the research report.
  • Input (required) — Choose between providing a single user question or a full Liner messages array as input.
  • Question (required) — The user question to send as the final message. Required if Input is set to 'Question'.
  • Messages JSON (required) — JSON-formatted array of conversation messages including roles and content, with the final message role being 'user'. Required if Input is set to 'Messages JSON'.
  • Include Raw Events — Boolean flag to include raw Server-Sent Event envelopes in the output for debugging or extended information.
  • Request ID — Optional client-supplied identifier that is echoed back by Liner for request tracking purposes.

Output shape

a JSON object representing the full deep research report including the generated text, reasoning, references, search steps, and optionally raw event data.

The output is a detailed structure parsed from server-sent events returned by Liner's API. If 'Include Raw Events' is enabled, raw event envelopes are included. Errors in the streaming response are handled and surfaced as node errors.

Links

Discussion