You are helping me build the following AI agent workflow.

## Goal
Personalized Sales Outreach: Scrapes verified B2B leads from Apollo, generates personalized cold emails via AI.

## Specification
- What it does: You save time finding prospects by automatically scraping verified B2B contacts from Apollo that match your ideal customer profile. The AI researches each lead to draft personalized cold emails for you, which are queued in Telegram for your final approval before sending via Gmail.
- Trigger: On a schedule (Scheduled · daily)
- Autonomy: You approve along the way
- Expected setup effort: about an afternoon
- Tools/services involved:
  - Apollo: Lead scraping and verification
  - OpenAI Chat Model: Enriching leads with research and generating emails
  - Gmail: Sending approved cold emails
  - Telegram: Human approval of email drafts

## Known pitfalls, handle each one explicitly in your implementation
1. Gmail sending quotas may be exceeded quickly, so batch emails and add delays between sends.
2. Telegram notification failures can cause lost approvals, so build a fallback queue or retry mechanism.

## Reference implementation
https://n8n.io/workflows/8523 (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-personalized-sales-outreach/ via usecasesforagents.com