You are helping me build the following AI agent workflow.

## Goal
Real Estate Listing Scraper: Scrapes real estate listings from paginated sites using AI extraction and saves unique properties to a Google Sheet.

## Specification
- What it does: You automatically gather real estate listings from paginated websites and save unique properties directly to your Google Sheet. This agent uses AI to extract key details like price, area, and amenities while preventing duplicates based on listing URLs.
- Trigger: Run manually (Manual · on demand)
- Autonomy: Fully hands-off
- Expected setup effort: about an afternoon
- Tools/services involved:
  - Scrapegraph AI: Web scraping and data extraction
  - Google Gemini Chat Model: AI-powered information extraction and validation
  - Google Sheets: Storing structured listing data with deduplication

## Known pitfalls, handle each one explicitly in your implementation
1. Websites often block automated requests, so implement random delays between scraping actions to avoid IP bans.
2. Google Sheets has row limits and API quotas, so batch insert operations instead of writing one row at a time.
3. Scraping structures change frequently, so add error handling that alerts you when the expected HTML selectors are missing.

## Reference implementation
https://n8n.io/workflows/13657 (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/real-estate-listing-scraper/ via usecasesforagents.com