LightRAG icon

LightRAG

Query and manage LightRAG with optional live progress in n8n Chat

Actions11

Query

AI-generated

Summary

Send a query to the LightRAG knowledge base and return its answer. Optionally stream retrieval progress, source excerpts, and answer text to an n8n Chat Trigger.

Inputs

  • Request Body (required) — JSON object for the LightRAG query; defaults to the complete current input item. It must include a non-empty query string of at least 3 characters. An optional mode must be one of local, global, hybrid, naive, mix, or bypass; other fields are passed through to LightRAG.
  • Stream Progress to Chat — Defaults to false. When enabled, calls LightRAG’s /query/stream endpoint and streams retrieval progress, source excerpts, and answer text to the chat. Requires an n8n Chat Trigger configured for Streaming Response and a LightRAG server supporting /query/stream and include_progress.

Output shape

A list of items, normally one per input item, each containing the LightRAG response object.

With streaming enabled, the response object contains the accumulated response text and may include references, llm_generated, and response_time. Errors stop execution unless Continue On Fail is enabled, in which case an item with an error field is returned.

Links

Discussion