n8n Content Generator for WordPress

This n8n workflow automates the process of generating and publishing content to WordPress, leveraging AI capabilities and Airtable for content management. It can be triggered manually or by new records in Airtable, allowing for flexible content creation and publishing workflows.

What it does

This workflow streamlines your content creation and publishing process through the following steps:

1. Triggers:
   • Manually: The workflow can be executed on demand using the "Manual Trigger" node.
   • Airtable: It can also be triggered automatically when a new record is added to a specified Airtable base and table.
2. Initial Data Processing:
   • Edit Fields (Set): Standardizes and prepares the incoming data from the trigger, ensuring consistency for subsequent AI processing.
3. AI Content Generation:
   • AI Agent: Utilizes an AI agent (LangChain) to generate content based on the input. This agent is configured with:
     • Anthropic Chat Model: Employs an Anthropic large language model for generating human-like text.
     • Structured Output Parser: Ensures the AI output adheres to a predefined JSON structure, making it easy to parse and use.
     • Tools: Provides the AI agent with external capabilities:
       • Calculator: Allows the AI to perform mathematical operations if needed for content generation.
       • Wikipedia: Enables the AI to fetch information from Wikipedia for research and content enrichment.
       • HTTP Request Tool: Gives the AI the ability to make HTTP requests, potentially for fetching data from other APIs or web services.
4. Content Review and Approval (Human-in-the-Loop):
   • Slack: Sends the generated content to a Slack channel for human review and approval.
   • Wait: Pauses the workflow execution, awaiting a response or approval from Slack.
   • Respond to Webhook: If the workflow was triggered by a webhook, this node sends a response back after the content is generated and reviewed.
   • If: Checks the approval status from Slack to determine the next steps.
5. Conditional Publishing:
   • WordPress (True branch): If the content is approved, it is automatically published to WordPress.
   • Slack (False branch): If the content is rejected, a notification is sent back to Slack.
6. Batch Processing:
   • Loop Over Items (Split in Batches): Processes multiple content items in batches, ensuring efficient handling of large datasets.
   • Merge: Combines data from different branches or iterations of the workflow.
   • Split Out: Separates aggregated data into individual items for further processing.
7. Additional Data Handling:
   • Airtable: Performs additional operations on Airtable, possibly updating records with the content status or published URLs.
   • HTTP Request: Makes general HTTP requests for various purposes within the workflow.
   • Code: Executes custom JavaScript code for advanced data manipulation or logic.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n instance: A running n8n instance.
• Airtable Account: With credentials configured in n8n.
• Anthropic API Key: For the Anthropic Chat Model.
• Slack Account: With credentials configured in n8n for sending notifications.
• WordPress Instance: With credentials configured in n8n for publishing content.
• OpenAI API Key: For the OpenAI node (if used, though Anthropic is explicitly configured in the AI Agent).

Setup/Usage

1. Import the workflow: Download the JSON file and import it into your n8n instance.
2. Configure Credentials:
   • Set up your Airtable, Anthropic, Slack, WordPress, and OpenAI credentials in n8n.
3. Airtable Trigger:
   • If using the Airtable trigger, specify your Base ID and Table Name in the "Airtable Trigger" node.
4. Edit Fields (Set):
   • Review and adjust the fields being set in the "Edit Fields (Set)" node to match your incoming data structure and desired output format.
5. AI Agent Configuration:
   • Ensure the "AI Agent" node is correctly configured with the "Anthropic Chat Model" and the necessary tools ("Calculator", "Wikipedia", "HTTP Request Tool").
   • Verify the "Structured Output Parser" schema matches the expected output from your AI agent.
6. Slack Integration:
   • In the "Slack" node for approval, specify the target Channel and customize the message content.
   • The "Wait" node will pause until a response is received from Slack. Ensure your Slack integration is set up to send a signal back to n8n for approval/rejection.
7. WordPress Publishing:
   • In the "Wordpress" node, configure the Resource (e.g., Post), Operation (e.g., Create), and map the generated content fields to the appropriate WordPress post fields (title, content, status, etc.).
8. Activate the workflow: Once configured, activate the workflow. You can test it using the "Manual Trigger" or by adding a new record to your specified Airtable table.