NaLU AI icon

NaLU AI

Semantic data validation and extraction for chatbots using NaLU AI.

Validate Handoff (Validate_handoff)

AI-generated

Summary

Detects whether a user's reply expresses intent to speak with a human agent, by sending the chatbot's question and the user's reply to the NaLU AI service for semantic analysis.

Inputs

  • agent_input (required) — The question or statement the chatbot sent to the user (e.g. "Qual o seu nome?").
  • user_input (required) — The user's reply to analyze for handoff intent.
  • language — Conversation language: English (en), Portuguese (Brazil) (pt-BR), or Spanish (es). Defaults to pt-BR.

Output shape

One output item per input item. Its json is the parsed response from the NaLU AI endpoint POST https://naluai.dev/v1/extract/handoff.

The node source doesn't define the response fields, so they depend on the NaLU AI API. Requires the naluAiApi credential. With Continue On Fail enabled, a failed item returns {error: message}. Otherwise the node throws a NodeOperationError. Each output item is paired to its input item.

Examples

Example 1: Route a chatbot conversation to a live agent when the user asks for one, even indirectly.

agent_input = {{$json.lastBotMessage}}, user_input = {{$json.userMessage}}, language = en. Follow with an IF node on the returned handoff result.

Example 2: Check Brazilian Portuguese customer replies for requests to talk to a person.

agent_input = "Como posso ajudar?", user_input = {{$json.text}}, language = pt-BR.

Links

Discussion