n8n Workflow: Ollama Qwen LLM Appian Task Manager with Postgres Memory

This n8n workflow demonstrates a powerful conversational AI agent that can manage Appian tasks using an Ollama Qwen LLM, enhanced with persistent chat memory stored in a PostgreSQL database. It provides a robust framework for building intelligent assistants capable of understanding user requests and maintaining context across interactions.

What it does

This workflow orchestrates the following key functionalities:

1. Listens for Chat Messages: It is triggered by incoming chat messages, acting as the entry point for user interaction with the AI agent.
2. Initializes/Retrieves Chat Memory: Before processing a new message, it interacts with a PostgreSQL database to retrieve the conversation history, ensuring the AI agent maintains context.
3. Processes with an AI Agent: The core of the workflow uses an AI Agent node (likely a LangChain agent) to interpret the user's message in the context of the chat history. This agent is configured to use the Ollama Chat Model for its language understanding and generation capabilities.
4. Generates AI Response: The AI Agent processes the input and memory to formulate a relevant and context-aware response.
5. Stores Updated Chat Memory: After generating a response, the updated conversation (including the new message and the AI's response) is saved back to the PostgreSQL database, preserving the chat history for future interactions.
6. Responds to Webhook: The final AI-generated response is sent back to the originating chat platform via a webhook, completing the conversational loop.
7. Edits Fields (Utility): A "Set" node is included, likely for utility purposes such as transforming or filtering data at some point in the workflow, though its specific configuration isn't detailed in the provided JSON.
8. Sticky Note (Documentation): A sticky note is present, serving as a documentation element within the workflow, potentially containing instructions or explanations.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance to import and execute the workflow.
• Ollama Server: An accessible Ollama server running the Qwen LLM (or another compatible chat model) for the "Ollama Chat Model" node.
• PostgreSQL Database: A PostgreSQL database instance configured for the "Postgres Chat Memory" node to store and retrieve conversation history. You will need the connection details (host, port, database name, user, password).
• Chat Platform Integration: A chat platform (e.g., Slack, Discord, custom application) that can send messages to an n8n webhook and receive responses.
• n8n LangChain Nodes: Ensure the @n8n/n8n-nodes-langchain package is installed in your n8n instance, as this workflow heavily relies on its nodes (AI Agent, Ollama Chat Model, Postgres Chat Memory, Chat Trigger).

Setup/Usage

1. Import the Workflow:
   • Download the provided JSON.
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the "Import from JSON" button and paste the workflow JSON.
2. Configure Credentials:
   • Ollama Chat Model: Configure the credentials for your Ollama server.
   • Postgres Chat Memory: Set up credentials for your PostgreSQL database, providing the necessary connection details.
3. Configure Webhook:
   • Activate the "Webhook" node and copy its URL.
   • Configure your chat platform to send messages to this webhook URL.
4. Configure Chat Trigger:
   • The "When chat message received" node (Chat Trigger) will automatically handle incoming chat messages from the platform configured to send to the Webhook.
5. Activate the Workflow:
   • Once all credentials and configurations are set, activate the workflow to start processing chat messages.

This workflow provides a solid foundation for building interactive AI assistants with persistent memory, making conversations more natural and effective.