You are helping me build the following AI agent workflow.

## Goal
Real-Time Sales Pipeline Analytics: Scheduled CRM queries feed an LLM to detect stalled deals or low conversion rates.

## Specification
- What it does: You get instant Slack alerts when your sales pipeline stalls or conversion rates drop, so you can act immediately on critical issues. Daily snapshots of this data are automatically archived in Google Sheets to help you track historical trends over time.
- Trigger: On a schedule (Scheduled · periodic)
- Autonomy: Fully hands-off
- Expected setup effort: about an afternoon
- Tools/services involved:
  - n8n: Pipeline orchestration
  - CRM API: Pulling pipeline data from HubSpot, Salesforce, or Pipedrive
  - OpenAI Chat Model: Anomaly detection and risk highlighting
  - Slack: Sending real-time alerts to reps and managers
  - Google Sheets: Archiving daily snapshots for trend analysis

## Known pitfalls, handle each one explicitly in your implementation
1. LLM token limits can truncate long deal descriptions so include only key fields like stage and value in the prompt context.
2. Slack message formatting errors can break alerts so validate the markdown structure before sending to the channel.

## Reference implementation
https://n8n.io/workflows/5974 (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-time-sales-pipeline-analytics/ via usecasesforagents.com