You are helping me build the following AI agent workflow.

## Goal
Real Estate Lead Qualification: New property inquiry form? The agent scores the lead by budget and timeline, then saves qualified prospects to Airtable and emails you.

## Specification
- What it does: Get new property inquiries automatically scored by your agent's AI for budget and timeline fit before you even see them. It extracts key details from forms to identify qualified prospects, then saves them directly to Airtable and emails you immediately so you can focus on closing deals.
- Trigger: Triggered by an event (Event · on form submission)
- Autonomy: Fully hands-off
- Expected setup effort: about an afternoon
- Tools/services involved:
  - OpenAI Chat Model: Analyze lead input and generate score
  - Information Extractor: Extract structured data from form
  - Airtable: Store qualified leads in CRM
  - Gmail: Send notification to agent

## Known pitfalls, handle each one explicitly in your implementation
1. Gmail API quotas are strict for bulk operations, so implement a delay or batch queue if processing many leads at once. Use a small sleep interval between emails to avoid throttling errors.
2. Airtable field type mismatches cause silent failures, so ensure the LLM output types exactly match Airtable's expected formats like number or date. Map string outputs to numbers explicitly in the code layer.
3. Pii leakage risks exist if lead data is logged unnecessarily, so scrub sensitive information from logs and restrict Airtable access permissions. Enable audit logging only for administrative actions.

## Reference implementation
https://n8n.io/workflows/5428 (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-real-estate-lead-qualification/ via usecasesforagents.com