Warm Up a Target Account on LinkedIn with an AI Agent (Before the Ask)

# Account Warmer — Loadout

```yaml
loadout: account-warmup
tier: apex                # Apex tier (or an active Apex trial)
mcp: https://apex.leadshark.io/mcp
```

**The job:** Company-level play: research a target account, engage its team's posts over days, and only then send the personalized connect — familiarity before the ask.

## Before you run — ask the user

Only ask what you don't already know — check the conversation, your memory,
and the account itself first. Confirm anything you inferred in one line
instead of re-interviewing.

1. **Which target account, and what's your offer to them?**
2. **Who at the account matters most?** (roles, seniority)
3. **What pace feels safe — comments per person per week?**

## Steps

1. **See** — call `enrich_company`: Pull the account: firmographics, the team, lookalikes worth adding to the list.
2. **See** — call `list_person_posts`: Read what the key people are posting.
3. **Act** — call `comment_on_post`: Engage the team's posts with substance over several days — paced, not burst.
4. **Act** — call `send_connection_request`: Send the personalized connect once warm, with approval.

## STOP — approval gate

Present everything you found and drafted BEFORE any write. Do not send,
schedule, publish, connect, or create automations until the user approves.
Edits beat re-drafts: apply their changes, don't start over.

## Never

- Write to LinkedIn without explicit approval this run.
- Invent facts, numbers, results, or people.
- Contact anyone on the do-not-engage list, or double-touch someone
  (check `list_actions` first).
- Retry around a refusal — report what was refused and why.

## Done means

- The user approved the drafts (aim for ≤2 edit rounds).
- Approved actions are queued or scheduled — verify via `list_actions`.
- Report back: what ran, what's staged, and where to monitor it.

## Why this stays account-safe

Deliberately slow: engagement is spread across days inside your caps — burst engagement is what gets accounts flagged. Every comment and connect is approval-gated and logged.

Using Apex, research {company}: enrich the account, identify the 3 most relevant people for {my offer}, and read their recent posts. Over the next week, draft one substantive comment per person per few days (queued for my approval, inside my daily caps). Once we've engaged twice with someone, draft a personalized connection request referencing the thread. Nothing sends without me.