Google Drive Capabilities
You can iterate through Google Drive folders, find specific files, extract data, and even create or copy files and documents directly in your Drive.
Folder Navigation: Automatically navigate through Drive folders to find and process specific files.
File Creation: Generate new Google Docs, Sheets, or folders as part of your workflows.
Data Extraction: Extract key terms, structured fields, or summaries from various document types.
File Management: Copy or move files to reorganize your Drive based on workflow requirements.
Scalable Workflows: Handle multiple folders or files simultaneously without manual intervention.
Common Use Cases
Invoice Processing: Navigate a folder of invoices to extract payment details and store them in a Google Sheet.
Contract Review: Iterate through contracts to extract key terms and save them in a structured format for easy analysis.
Meeting Documentation: Process meeting notes or agendas to summarize action items and distribute organized reports.
Folder Organization: Automatically sort, copy, or archive files based on specific criteria, such as document type or date.
Tips and Notes
Structured Outputs: Use Lutra to generate Google Sheets or Docs that summarize or organize extracted data for easy access.
Folder Depth: Lutra supports nested folders, ask it to recursively look over folders and it can do so.
File Types: While optimized for PDFs and Google Docs, Lutra can also process other common file types like Excel. It can also produce PDF files and upload them to your Drive.
How to use Lutra
Lutra will automatically ask for account permissions when it needs to connect to them, you do not need to configure the integrations beforehand. To see what accounts have been connected to Lutra, you can visit the integrations page.
Use the 📎 attach button to provide a Google Drive folder for Lutra to access.
Select the folder in the picker that comes up.
Note: When using the picker to select the folder, be careful not to double-click to enter the folder you're selecting instead. Select the folder (single-click).
Send Lutra a message to make sure it can view the folder contents, for example, you can ask it to:
"List the files in this folder"
Ask it to process a few of the files to test.
"Process the first two files and extract X, Y, Z from them. Put the output into a Google sheet"
Check and refine the behavior of the AI by giving it feedback about the data it extracted. Test the process on more files, and once it works, you can scale it up.
Create a playbook to automate and scale up this process.
Tips for best results
Start Small: It’s always a good idea to test Lutra on a small number of examples before scaling up. Start with a limited dataset or a few PDF files to ensure the process runs smoothly.
Provide Feedback: Lutra uses AI to adapt and improve based on your feedback. If the initial results don’t meet your expectations, let Lutra know. It can fine-tune its approach, improving both the data it extracts and how it structures the output.
Provide feedback directly, for example:
"You missed extracting the invoice due dates, please try again."
Scale and Automate: Once you’re happy with the results, you can scale up the process by asking Lutra to create playbooks to run on many more documents, or on a scheduled basis.
You can also share the playbook with your team so that they can reuse the same automation.
Review Details: You can review the execution logs to see all the details of how Lutra performs tasks. Click on the log entries in the Chat view to expand them to see details.