LINE Plant Care AI Bot: Photo Analysis with n8n + Gemini

Turn your LINE Official Account into an AI plant care assistant that understands photos—using n8n + Google Gemini

This n8n workflow powers a LINE-based plant care bot that can analyze plant images, identify the plant when possible, diagnose visible health symptoms, and provide care advice—then continue the conversation with context and saved history.

What this workflow does

• Receives LINE messages via a Messaging API webhook and routes requests based on whether the user sends an image or text.
• For image messages:
  • Downloads the image from the LINE content API.
  • Loads the user’s active plant, preferred language, and recent conversation history from n8n Data Tables.
  • Sends the image plus context to Google Gemini to identify the plant (or defer when uncertain) and assess visible symptoms, possible causes, and care guidance.
  • If the image matches the active plant, it keeps the identity; if it’s a new plant, it upserts a plant profile and updates the user’s active plant; if uncertain, it asks for a better diagnostic photo without changing the active plant.
  • Stores the result as a plant observation in an n8n Data Table and replies in LINE with a formatted summary and confidence.
• For text messages: loads the user’s active plant and recent observations, then asks Google Gemini for a contextual reply in LINE.
• Conversation memory: saves user and assistant messages to conversation history, updating the saved preferred language only when the user explicitly requests a language change.

Use cases

• Give customers plant care support inside LINE—without building a separate frontend.
• Help users diagnose issues from leaf spots, discoloration, or other visible symptoms.
• Maintain per-user plant profiles and observation history to improve follow-up answers.

Technical details

• n8n nodes/workflow logic: webhook, if, set, switch, merge, aggregate.
• Integrations: LINE Messaging API webhook + LINE content API, Google Gemini.
• Data storage: n8n Data Tables for plant profiles, observations, conversation history, and language preference.