You are helping me build the following AI agent workflow.

## Goal
Tax Compliance and Revenue Anomaly Detection: Monitors revenue streams for tax anomalies using triple AI validation.

## Specification
- What it does: Eliminate hours of manual tax review by automatically monitoring revenue streams for anomalies and ensuring regulatory compliance across multiple data sources. The system fetches income data, applies triple AI validation to detect errors, syncs results back to your accounting software, and delivers instant alerts via WhatsApp along with detailed email reports.
- Trigger: Triggered by an event (Scheduled · continuous monitoring)
- Autonomy: Fully hands-off
- Expected setup effort: a few focused days
- Tools/services involved:
  - MagicCSV: Fetching revenue data from accounting software
  - Anthropic Chat Model: Categorization, anomaly detection, and compliance verification
  - Gmail: Sending tax summary reports
  - WhatsApp Business Cloud: Delivering instant compliance alerts

## Known pitfalls, handle each one explicitly in your implementation
1. Tax regulations vary by jurisdiction so you must parameterize your LLM prompts with specific regional rules for each user.
2. WhatsApp API quotas are strict so implement a cooldown mechanism to prevent message throttling during high-volume runs.
3. MagicCSV data formats may shift if the source accounting software updates its schema, so add validation checks before processing.

## Reference implementation
https://n8n.io/workflows/12790 (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/automated-tax-compliance-and-revenue-anomaly-detection/ via usecasesforagents.com