Skip to content

AI Customer Care Chat Assistant

A conversational AI workflow that provides customer support through a chat interface, powered by OpenRouter's language models with conversation memory to maintain context across interactions.

Purpose

No business context provided yet — add a context.md to enrich this documentation.

How It Works

  1. Chat Interface: The workflow exposes a public chat interface that customers can access to ask questions or request support
  2. AI Processing: When a message is received, an AI agent processes the request using a helpful customer care persona
  3. Language Model: OpenRouter's chat model generates appropriate responses based on the customer's inquiry
  4. Memory Management: The system maintains conversation history using a sliding window of the last 10 exchanges to provide contextual responses
  5. Response Delivery: The AI agent sends the generated response back to the customer through the chat interface

Workflow Diagram

graph TD
    A[When chat message received] --> B[AI Agent]
    C[OpenRouter Chat Model] --> B
    D[Simple Memory] --> B

    style A fill:#e1f5fe
    style B fill:#f3e5f5
    style C fill:#fff3e0
    style D fill:#e8f5e8

Trigger

Chat Trigger: A public chat interface that activates when users send messages. The trigger is configured to be available in chat applications and accepts public interactions.

Nodes Used

Node Type Purpose
Chat Trigger Receives incoming chat messages from users and initiates the workflow
AI Agent Orchestrates the conversation flow and manages interactions between components
OpenRouter Chat Model Provides the language model capabilities for generating responses
Simple Memory Maintains conversation context using a buffer window of recent exchanges

External Services & Credentials Required

  • OpenRouter API: Requires an OpenRouter account and API credentials
    • Credential name: "OpenRouter Org"
    • Used for: Accessing language models through OpenRouter's API

Environment Variables

No specific environment variables are configured in this workflow. All settings are managed through node parameters and credentials.

Data Flow

Input: - User chat messages (text) - Conversation history from memory buffer

Processing: - AI agent processes messages with customer care context - OpenRouter model generates appropriate responses - Memory system updates conversation history

Output: - AI-generated responses to user queries - Updated conversation memory for future interactions

Error Handling

No explicit error handling nodes are present in this workflow. The system relies on n8n's default error handling mechanisms and the built-in error management of the AI agent node.

Known Limitations

  • Memory is limited to the last 10 conversation exchanges
  • No explicit error handling for API failures
  • Workflow is currently inactive (not deployed)
  • No custom tools or knowledge base integration

No related workflows specified in the current context.

Setup Instructions

  1. Import Workflow: Import the workflow JSON into your n8n instance

  2. Configure OpenRouter Credentials:

    • Create an OpenRouter account at openrouter.ai
    • Generate an API key
    • In n8n, create new credentials of type "OpenRouter API"
    • Name the credential "OpenRouter Org"
    • Enter your API key

  3. Configure Chat Trigger:

    • The trigger is already set to public access
    • Note the webhook URL generated for the chat interface
    • Optionally customize the chat interface settings

  4. Adjust AI Agent Settings:

    • Review the system message: "You are a helpful customer care assistant"
    • Modify the persona or instructions as needed for your use case

  5. Configure Memory Settings:

    • Current setting maintains 10 conversation exchanges
    • Adjust the context window length if needed

  6. Test the Workflow:

    • Activate the workflow
    • Access the chat interface using the webhook URL
    • Send test messages to verify responses

  7. Deploy:

    • Ensure the workflow is activated
    • Share the chat interface URL with users
    • Monitor performance and adjust settings as needed