n8n Workflow: Extract and Process Information from PDF using Claude and Gemini (Partial Workflow)

This n8n workflow demonstrates a foundational structure for extracting and processing information, likely from PDF files, using various n8n core functionalities. While the full integration with AI models like Claude or Gemini is not explicitly defined in this JSON, the workflow provides the initial steps for file handling and data manipulation.

What it does

This workflow outlines the following steps:

1. Manual Trigger: Initiates the workflow manually, allowing for on-demand execution.
2. Google Drive: This node is present, suggesting an intention to interact with Google Drive, potentially to retrieve or store files.
3. Extract from File: This node is designed to extract content from a file, which could be a PDF, CSV, or other supported formats. This is a crucial step for getting data out of a document.
4. Edit Fields (Set): This node allows for the transformation and manipulation of data that has been extracted or received, enabling you to set, modify, or remove fields.
5. HTTP Request: This node is included, indicating a capability to make external API calls. This is where the integration with AI models like Claude or Gemini would typically occur, by sending the extracted text for processing and receiving the AI's response.
6. Sticky Note: A sticky note is present, likely for documentation or to leave a comment within the workflow.

Prerequisites/Requirements

• n8n Instance: A running instance of n8n.
• Google Drive Account (Optional): If you intend to use the Google Drive node to fetch or store files, you will need a Google Drive account and corresponding n8n credentials.
• API Keys/Credentials for AI Services (Not in JSON, but implied by directory name): To fully realize the "Claude and Gemini" aspect mentioned in the directory name, you would need API keys and credentials for these respective AI services (e.g., Anthropic Claude, Google Gemini). These would be configured within the HTTP Request node or dedicated AI nodes if available.

Setup/Usage

1. Import the Workflow:
   • Save the provided JSON content to a file (e.g., workflow.json).
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the "Import from JSON" button and select your workflow.json file.
2. Configure Nodes:
   • Google Drive (if used): If you plan to use the Google Drive node, configure your Google Drive credentials.
   • Extract from File: Ensure this node is correctly configured to handle the type of file you intend to process (e.g., PDF, CSV). You might need to provide binary data from a previous node or specify a file path.
   • HTTP Request (for AI integration): If you want to integrate with Claude or Gemini, you would need to configure this node with the appropriate API endpoint, authentication (e.g., API key in headers), and request body containing the extracted text.
3. Activate the Workflow: Once configured, activate the workflow by toggling the "Active" switch in the top right corner.
4. Execute:
   • For testing, click "Execute Workflow" on the "Manual Trigger" node.
   • For a fully automated scenario, you would typically replace the "Manual Trigger" with a different trigger node (e.g., Webhook, Cron, Google Drive Watcher) that responds to new file uploads or events.

This workflow provides a robust starting point for building complex document processing automations.