AI Data Extraction with Dynamic Prompts and Airtable

This n8n workflow automates the process of extracting specific data points from text content using AI, dynamically generating prompts, and storing the extracted information in Airtable. It's designed to streamline data processing tasks where consistent, structured data needs to be pulled from unstructured text.

What it does

This workflow performs the following key steps:

1. Triggers Manually: The workflow is initiated manually, allowing for on-demand processing.
2. Extracts Text from File (Optional): If a file is provided (e.g., PDF, TXT), it extracts the raw text content.
3. Prepares AI Prompt: It constructs a dynamic prompt for the AI model, including the text to be analyzed and instructions for the desired data extraction.
4. Generates AI Response: It sends the prepared prompt to an OpenAI Chat Model to perform the data extraction.
5. Processes AI Output: It takes the AI's response and attempts to parse the extracted data (presumably JSON).
6. Filters for Valid Data: It checks if the AI successfully extracted the required data.
7. Updates Airtable (if valid): If valid data is extracted, it updates a record in Airtable with the new information.
8. Handles Invalid Data: If the AI fails to extract valid data, it logs this and proceeds without updating Airtable.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance.
• Airtable Account & Base: An Airtable account with a base and table configured to store your extracted data. You'll need your Airtable API Key, Base ID, and Table Name.
• OpenAI API Key: An OpenAI API key with access to their chat models (e.g., GPT-3.5, GPT-4).
• Optional: File Input: If you intend to extract data from files, ensure your n8n instance can access and process the file types you'll be using.

Setup/Usage

1. Import the Workflow:

   • Download the provided JSON file.
   • In your n8n instance, click "Workflows" in the left sidebar.
   • Click "New" and then "Import from JSON".
   • Paste the JSON content or upload the file.

2. Configure Credentials:

   • Airtable: Locate the "Airtable" node. Click on the credential field and either select an existing Airtable credential or create a new one. You'll need your Airtable API Key.
   • OpenAI Chat Model: Locate the "OpenAI Chat Model" node. Click on the credential field and either select an existing OpenAI credential or create a new one. You'll need your OpenAI API Key.

3. Customize Airtable Node:

   • In the "Airtable" node, configure the Base ID and Table Name to match your Airtable setup.
   • Ensure the Record ID and Fields being updated correspond to the data you expect to extract from the AI.

4. Customize AI Prompt (Code Node):

   • Open the "Code" node named "Prepare AI Prompt".
   • Review the JavaScript code. This is where the dynamic prompt is constructed. You will likely need to adjust the prompt variable to instruct the AI on what data to extract and in what format (e.g., JSON).
   • The current setup seems to expect a text field from previous nodes and a promptTemplate from the Airtable node. Ensure these match your actual data flow or adjust accordingly.

5. Test and Activate:

   • Run the workflow once using the "Execute Workflow" button on the "Manual Trigger" node to test the setup.
   • Verify that the AI extracts the data correctly and that Airtable is updated as expected.
   • Once confident, activate the workflow by toggling the "Active" switch in the top right corner.

This workflow provides a powerful foundation for automating AI-driven data extraction and integration with Airtable, allowing you to adapt it to various use cases by modifying the AI prompt and Airtable mapping.