---
name: claudeemailskill
description: Use when Codex, Hermes, OpenClaw, Claude Code, Cowork, or another AI agent needs to plan, review, implement, audit, or improve email work focused on Claude-oriented email prompts, reviews, and campaign drafting. Triggers include requests about Prompt packs, Human-in-the-loop review, Campaign draft evaluation, email briefs, email QA, subscriber journeys, production handoffs, approval gates, and agent-safe email operations.
---

# Claude Email Skill

Use this skill to turn email work into an agent-safe operating routine. Start by identifying whether the task is analysis, drafting, implementation planning, QA, or production execution. Keep production execution behind explicit human approval.

## Workflow

1. Clarify the email surface: campaign, newsletter, automation, transactional message, template, deliverability issue, or reporting task.
2. Ask for source material that changes the answer: audience, lifecycle stage, offer, sending platform, segment rules, screenshots or exports, recent metrics, brand voice, compliance constraints, and existing copy.
3. Produce structured output with these headings when relevant: assumptions, missing inputs, recommendation, draft or implementation notes, QA checks, risk level, and approval needed.
4. Do not send, import contacts, alter suppression rules, change DNS, or modify production automations without explicit confirmation.
5. If the task is complex or high risk, read references/operating-checklist.md before finalizing.

## Review Criteria

- The recommendation matches Claude-oriented email prompts, reviews, and campaign drafting.
- The audience and permission basis are clear.
- The message promise, segment, timing, and call to action are consistent.
- Deliverability, unsubscribe, complaint, and compliance risks are surfaced.
- Any live-system action is separated from analysis or drafting.

## Output Pattern

Return concise, executable artifacts: briefs, checklists, diffs, campaign outlines, sequence maps, QA tables, or post-send retrospectives. Prefer concrete next actions over generic best practices.