n8n AI Agent Workflow: Chat-Driven Content Generation with OpenAI and Anthropic

This n8n workflow demonstrates a powerful AI agent setup that can respond to chat messages, leverage different large language models (LLMs) like OpenAI's GPT and Anthropic's Claude, and maintain conversational memory. It provides a flexible foundation for building sophisticated AI-driven applications, such as content generation, customer support, or interactive assistants.

What it does

This workflow sets up an n8n AI Agent that can receive chat messages, process them using either OpenAI's or Anthropic's chat models, and maintain a conversational memory. While the provided JSON defines the core components, it's designed to be extended with specific tools and logic to perform tasks like generating blog content, answering questions, or interacting with other services.

Specifically, the workflow defines the following components:

1. Chat Trigger: Listens for incoming chat messages to initiate the workflow.
2. AI Agent: The central orchestrator that uses language models and tools to respond to queries.
3. OpenAI Chat Model: Integrates with OpenAI's large language models (e.g., GPT-4) for generating responses.
4. Anthropic Chat Model: Integrates with Anthropic's Claude models for generating responses.
5. Simple Memory: Maintains a buffer of recent conversational turns, allowing the AI agent to remember context within a conversation.
6. AI Agent Tool: A placeholder for defining specific tools that the AI agent can use (e.g., a tool to publish to Ghost CMS, a tool to search the web, etc.).
7. Sticky Note: A helper node for documentation or temporary notes within the workflow.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance (self-hosted or cloud).
• OpenAI API Key: For the OpenAI Chat Model node.
• Anthropic API Key: For the Anthropic Chat Model node.
• LangChain Credentials: While not explicitly shown as a separate credential node, the LangChain nodes will require appropriate API keys for the chosen LLMs.

Setup/Usage

1. Import the Workflow:
   • Download the provided JSON file.
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the "..." menu in the top right and select "Import from JSON".
   • Paste the JSON content or upload the file.
2. Configure Credentials:
   • Locate the "OpenAI Chat Model" and "Anthropic Chat Model" nodes.
   • Configure the respective API keys for OpenAI and Anthropic in their credential fields.
3. Connect to a Chat Service:
   • The "When chat message received" trigger needs to be connected to a chat service (e.g., Slack, Telegram, Discord, custom webhook). Configure this node to listen for messages from your desired platform.
4. Define AI Agent Logic and Tools:
   • The "AI Agent" node is the core. You will need to configure its prompt, the language model it should use (either OpenAI or Anthropic, or both with conditional logic), and any specific "Tools" it should have access to.
   • The "AI Agent Tool" node is a placeholder. To make the agent perform specific tasks (like publishing to Ghost CMS as hinted by the directory name), you would configure this node (or add more AI Agent Tool nodes) to define functions the agent can call. For example, a tool could be an HTTP Request node to interact with the Ghost CMS API.
5. Activate the Workflow:
   • Once configured, activate the workflow by toggling the "Active" switch in the top right corner of the workflow editor.

This workflow provides a robust starting point for building intelligent, chat-driven automation with n8n and leading AI models.