Liner icon

Liner

Search the web and generate source-backed answers with Liner

Actions7

Answer → Quick Answer

AI-generated

Summary

Generate a fast, source-backed answer to a user query using the Liner platform.

Inputs

  • lang — Optional language code (e.g., en, ko, ja) to specify the answer language.
  • inputMode (required) — Select whether to input a single user question or a full Liner messages JSON array (conversation history).
  • question (required) — The question string to send as the final user message, required if 'Input' is set to 'Question'.
  • messagesJson (required) — A JSON array of messages with 'role' ('user' or 'assistant') and 'content' fields representing conversation history; the final message must have role 'user'. Required if 'Input' is set to 'Messages JSON'.
  • includeEvents — Boolean flag to include raw Server-Sent Event envelopes in the output.
  • requestId — Optional client-supplied ID that the Liner API will echo back in the response.

Output shape

a JSON object containing the source-backed answer and related metadata returned by Liner's /quick-answer endpoint

The output includes text answer, reasoning, references, and optionally raw server-sent event data if 'Include Raw Events' is true. Errors are returned as JSON objects with error messages. If multiple input items are processed, outputs are paired by input index.

Examples

Example 1: Provide a single question to get a quick, source-backed answer

Set 'Input' to 'Question' and fill in the 'Question' field with the user's query.

Example 2: Provide full conversation context using messages array

Set 'Input' to 'Messages JSON' and provide an array of message objects representing conversation history, with the last message from user.

Links

Discussion