n8n Chat Trigger + DeepSeek LangChain Agent Workflow

Turn chat messages into DeepSeek-powered AI replies in n8n

This n8n workflow lets you receive a user’s message via an n8n public chat trigger, then forwards it to a LangChain AI Agent using the DeepSeek Chat Model (deepseek-flash) to generate a response—perfect for embedding conversational AI into your web widget or app.

What this workflow does

• Receives a chat message through an n8n chat trigger (public chat/webhook URL).
• Normalizes the payload by mapping the incoming message into a chatInput field.
• Sends chatInput to a LangChain AI Agent for processing and response generation.
• Uses the DeepSeek Chat Model (deepseek-flash) to produce the agent’s reply.

Setup (quick and practical)

• Create and select a DeepSeek API credential for the DeepSeek Chat Model node.
• Enable the chat trigger and copy the public chat/webhook URL to the client or widget that will send chat messages to n8n.
• (Optional) Update the agent instructions/prompt inside the AI Agent node to match the behavior you want.

Use cases

• Customer support chat where users ask questions and receive DeepSeek-generated answers via an n8n-powered endpoint.
• SaaS in-app assistant for contextual Q&A, where your UI sends messages to the n8n Chat Trigger and displays the response.
• Automation engineers prototyping conversational workflows using n8n + LangChain without building a full backend.

Technical details

• n8n Chat Trigger: n8nn8n-nodes-langchainchat trigger
• Payload mapping: set node to map incoming message into chatInput
• LangChain Agent: n8nn8n-nodes-langchainagent
• DeepSeek model: n8nn8n-nodes-langchainlm chat deep seek using deepseek-flash