n8n Workflow: Audit Interview Feedback Report via Slack with GPT-4o Mini and Google Sheets

This n8n workflow automates the process of generating and reporting audit interview feedback. It reads interview data from a Google Sheet, uses an AI model (GPT-4o Mini via Azure OpenAI) to generate a summary, and then posts this summary to a Slack channel. The workflow includes conditional logic to only process "Completed" interviews.

What it does

This workflow streamlines the reporting of audit interview feedback by:

1. Triggering Manually: The workflow is initiated manually.
2. Reading Google Sheet Data: It fetches all rows from a specified Google Sheet.
3. Filtering Completed Interviews: It checks each row to ensure the interview status is "Completed". Only completed interviews proceed.
4. Preparing Data for AI: For each completed interview, it structures the relevant feedback data into a format suitable for an AI prompt.
5. Generating AI Summary: It uses an Azure OpenAI Chat Model (configured for GPT-4o Mini) with a "Basic LLM Chain" to generate a concise summary of the interview feedback based on a predefined prompt.
6. Posting to Slack: The generated AI summary, along with key interview details, is then posted as a message to a designated Slack channel.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance.
• Google Sheets Account: Access to a Google Sheet containing your audit interview feedback.
• Azure OpenAI Service: An Azure OpenAI account with access to the GPT-4o Mini model.
• Slack Account: A Slack workspace and a channel where you want to post the reports.
• n8n Credentials: Configured credentials for:
  • Google Sheets (OAuth 2.0 or Service Account)
  • Azure OpenAI Chat Model (API Key)
  • Slack (OAuth 2.0 or Bot Token)

Setup/Usage

1. Import the Workflow:
   • Copy the provided JSON code.
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the "Import from JSON" button and paste the workflow JSON.
2. Configure Google Sheets Node (Node ID: 18):
   • Select your Google Sheets credential.
   • Specify the "Spreadsheet ID" and "Sheet Name" where your interview feedback data is located.
   • Ensure the "Operation" is set to "Get All".
3. Configure If Node (Node ID: 20):
   • This node is pre-configured to check if the Status column in your Google Sheet is equal to "Completed". Adjust the column name (Status) if it's different in your sheet.
4. Configure Code Node (Node ID: 834):
   • This node prepares the prompt for the AI. Review the JavaScript code to ensure it correctly extracts the relevant fields from your Google Sheet data (e.g., Candidate Name, Interviewer, Feedback, Rating, etc.) and formats them into a clear prompt for the AI.
5. Configure Basic LLM Chain Node (Node ID: 1123):
   • This node orchestrates the AI interaction.
   • Prompt: Ensure the prompt template is suitable for generating an audit interview feedback summary. It currently uses the output from the "Code" node.
6. Configure Azure OpenAI Chat Model Node (Node ID: 1253):
   • Select your Azure OpenAI credential.
   • Specify the "Deployment Name" for your GPT-4o Mini model.
   • Adjust other parameters like "Temperature" or "Max Tokens" as needed for your desired output.
7. Configure Slack Node (Node ID: 40):
   • Select your Slack credential.
   • Specify the "Channel" where the feedback reports should be posted.
   • Customize the "Text" field to format the Slack message with the AI-generated summary and other relevant details from the Google Sheet.
8. Activate the Workflow: Once all configurations are complete, activate the workflow.
9. Execute the Workflow: Click "Execute Workflow" on the "Manual Trigger" node to run it.

The workflow will then fetch data, process completed interviews, generate AI summaries, and post them to your Slack channel.