You are helping me build the following AI agent workflow.

## Goal
Vehicle Insurance Claim Approvals: Monitors for vehicle insurance claim emails, extracts PDF data via OCR, evaluates against underwriting rules.

## Specification
- What it does: You get instant email approvals for vehicle insurance claims as soon as your agent finds the PDF attachments in Gmail. It automatically checks for duplicates, extracts data via OCR, evaluates the case against underwriting rules, and sends you signed approval or rejection letters directly to your inbox.
- Trigger: Triggered by an event (Event · on new email with PDF attachment)
- Autonomy: Fully hands-off
- Expected setup effort: about an afternoon
- Tools/services involved:
  - Gmail: Monitor claims inbox and send decision emails
  - Google Sheets: Log claims and detect duplicates
  - PDF API Hub: Extract text from PDFs and generate signed letters
  - OpenAI Chat Model: Evaluate claims and determine approval status

## Known pitfalls, handle each one explicitly in your implementation
1. OCR accuracy drops on low-resolution scans so implement a validation step to reject blurry images before processing.
2. Email attachments may exceed Gmail size limits for direct parsing, so use temporary storage links if files are large.

## Reference implementation
https://n8n.io/workflows/15232 (workflow template)
Fetch and inspect this before building. If it matches my stack, adapt it;
if not, rebuild the pattern with my tools.

## Process requirements
1. Before building: ask me which of the listed tools I actually use and
   what my platform is (n8n / Make / code / other). Do not assume.
2. Adapt the pattern to my answers; do not force the reference stack.
3. Address every pitfall above; tell me how you handled each.
4. Provide a test plan I can run before letting this touch real data.
5. Ask before any step that sends messages, modifies data, or spends money.

Source: https://usecasesforagents.com/use-case/automate-vehicle-insurance-claim-approvals/ via usecasesforagents.com