send-outreach

Files (1)

• SKILL.md

SKILL.md (first lines)

## Browser Preflight (MANDATORY — do this FIRST)

Before delegating to any agent, attach the Claude Chrome extension to this session:

1. Tell the user in chat:
   > "I need browser access to LinkedIn. You may see a permission prompt — please approve it. I'll wait."
2. Navigate to `https://www.linkedin.com/feed/` using the Browser tool. This triggers Claude's permission system and attaches the Chrome extension to the session.
3. Confirm the LinkedIn feed page loaded (content visible, not an error/login wall). If it fails, tell the user the extension isn't connected and stop — do NOT delegate to the agent.

If LinkedIn already loaded earlier in this session, this step passes instantly.

Delegate to the **outreach** agent in **send mode** with these constraints:

- Re-inspect each lead's LinkedIn thread before sending (see **Pre-Send Thread Validation** in the outreach agent). Mark any draft as STALE and skip if the thread has changed since the draft was created.
- Read pre-drafted messages from outreach agent memory (status = "Ready to Send")
- If no drafts found, tell the user: "No drafted messages found. Run /gtm-kit:prep-outreach first."
- Present each draft or a clearly enumerated batch and wait for explicit user approval before sending
- Before sending, confirm each draft has a canonical dashboard `lead_id` resolved from the matched lead (`id`, `_id.$oid`, then `original_id`); never send or sync a draft whose `lead_id` is only a LinkedIn slug
- Send each message on LinkedIn using the appropriate method (DM, InMail, or connect note)
- Batch only messages approved together by the user
- Update status in agent memory after each send (Sent / Skipped / Error / Paused)
- Do NOT re-research prospects — use existing drafts as-is
- If LinkedIn shows rate limit warning, stop immediately

After completing this command, sync all data per `references/storage.md`.

From Kasper-AI/gtm-kit-plugin, MIT.

More from Kasper-AI/gtm-kit-plugin