Skip to main content

This is a prototype vision of how a future government service could work. It's not a real service yet, but we're exploring what it could look like. Your feedback will help shape the real service.

Step 1: Access Document Processing Interface - Planning AI Walkthrough

Find and open your deployed AI document extraction interface

Walkthrough progress

Step 1 of 4 • 2 minutes

Step 1 2 minutes

Access the Document Processing Interface

Find your AI document extraction service URL from the CloudFormation stack outputs.

Find your Planning AI application URL in CloudFormation Outputs
The application is ready to process planning documents

Expected outcome

  • CloudFormation stack shows CREATE_COMPLETE status
  • ProcessingURL output is visible in the Outputs tab
  • Document processing interface opens in a new browser tab

How to find your processing interface URL

  1. Open the CloudFormation console

    Go to CloudFormation console in us-east-1 (N. Virginia) (opens in new tab)

  2. Find your stack

    Look for a stack named ndx-try-planning-ai-[timestamp]

    The status should be CREATE_COMPLETE

  3. Click on the stack name

    This opens the stack details page

  4. Go to the "Outputs" tab

    You'll see a list of outputs from the deployment

  5. Find the ProcessingURL output

    Look for the row with Key: ProcessingURL

    The Value will be a URL like: https://[random-id].execute-api.us-east-1.amazonaws.com/prod/extract

  6. Click the ProcessingURL value

    This opens the document processing interface in a new tab

Tip: Bookmark this URL or keep the tab open - you'll need it for the rest of the walkthrough.

What you should see

When you open the processing URL, you should see:

  • A clean web interface with "Planning Application AI - Document Extraction" heading
  • Upload area with drag-and-drop zone or file browser button
  • Supported formats notice: "PDF, JPEG, PNG (max 10MB)"
  • Sample documents section with 3 downloadable applications
  • Empty results area ready to show extracted fields

Troubleshooting

Stack not showing CREATE_COMPLETE

If your stack shows CREATE_IN_PROGRESS:

  • Wait 3-5 minutes for deployment to complete
  • Refresh the CloudFormation console page
  • Check the "Events" tab to see deployment progress
  • Textract and S3 setup typically takes 2-3 minutes

If your stack shows CREATE_FAILED or ROLLBACK_COMPLETE:

  • Check the "Events" tab to see which resource failed
  • Common issues: IAM permissions for Textract, S3 bucket name conflicts
  • Try deleting the stack and redeploying
  • Contact support if the issue persists
ProcessingURL not in Outputs

If you don't see ProcessingURL in the Outputs tab:

  • Check the stack is fully deployed (CREATE_COMPLETE status)
  • Refresh the page - outputs may take a moment to appear
  • Check you're looking at the correct stack (ndx-try-planning-ai-...)
  • Try the "Resources" tab and find the API Gateway - the invoke URL is your processing URL
  • Ensure API Gateway Lambda integration completed successfully
Processing interface shows error

If the processing interface page shows an error or won't load:

  • Check you're using the correct region (us-east-1)
  • Check your browser isn't blocking the page (HTTPS should work)
  • Try opening in an incognito/private window
  • Check the stack Events tab for any errors during API Gateway or Lambda setup
  • Wait a few minutes - API Gateway endpoints can take 1-2 minutes to activate after deployment
  • Verify Textract service is available in us-east-1 region
Sample documents not showing

If the sample documents section is empty:

  • Wait 1-2 minutes - sample documents are uploaded during stack creation
  • Refresh the processing interface page
  • Check the S3 bucket named ndx-try-planning-ai-[timestamp]-documents contains files
  • If issues persist, contact support
Note Screenshot placeholder: In production, this page would include annotated screenshots showing the CloudFormation Outputs tab and the document processing interface.