Marketing Email

@./format-guidelines.md Read when: starting this skill @../_shared/storytelling-frameworks.md Read when: selecting a narrative framework in Step 5 @../_shared/persuasion-principles.md Read when: selecting persuasion principles in Step 6

Email Skill

You are helping the owner produce a complete, publishable marketing or sales email. The email must comply with the format guidelines in this skill folder, reflect the company's brand voice, and — if a campaign briefing is present — serve the briefing's goals.

This skill covers marketing email (bulk campaigns, newsletters, nurture sequences) and sales email (cold or warm 1:1 outreach) as two flavors of the same discipline. It does not cover everyday personal or internal professional correspondence. This skill is language-, industry-, and audience-neutral. It works for any output language and any B2B or B2C context. Calibration happens through the loaded context files, the campaign briefing (if any), and the owner's topic input.

Step 0: Recall learnings

If .claude/learnings.md exists, read it silently. Apply all relevant entries to inform this run. Do not announce this step. If the file is absent, continue normally.

Step 1: Load context

Check for a ## Context files table in CLAUDE.md via:

grep -A 200 '## Context files' CLAUDE.md 2>/dev/null || echo "(no context table)"

CLAUDE.md files may exist at multiple hierarchy levels (workspace root, project root, sub-directory) and the harness loads all applicable ones. Where multiple tables exist, rows from more specific CLAUDE.md files take precedence.

Do not enumerate required categories by name. Context rows carry free-form, user-chosen labels — there is no schema of category names to match against. Instead, Read every file listed in the File column, then assess each row's Summary to work out what it covers, and map the loaded files to this skill's content needs:

Where multiple files plausibly cover the same need, pick the one whose Summary best fits the task at hand and note the choice.

Warn on semantic absence, not label absence — "No brand voice context found", never "no writing-style row". Gate on two needs only: if no loaded file plausibly covers brand voice, or none covers organization background, ask once whether this is intentional or whether the user should pause and run /content-onboarding.

"I don't see any [writing style / organization background] context. Is this intentional, or should I pause while you run /content-onboarding?"

• Intentional: note the gap; label the final output ⚠ DEGRADED OUTPUT — no [brand voice / organization background] context
• Pause: direct the owner to onboarding and stop.

For absent audience, language, or email-specific rules: note silently and continue — never ask.

See ../_shared/context-categories.md for the full guide to common context patterns and the authoring rules this step follows.

Step 2: Check for campaign briefing

Determine the briefing path:

1. If the owner passed a file path as an argument ($ARGUMENTS), use that path.
2. Otherwise, check for brief.md in the current working directory:

ls brief.md 2>/dev/null && echo "found" || echo "missing"

• Found (either via argument or brief.md): read the file and note its key messages, goals, and constraints. Confirm: "✓ Campaign briefing loaded from <path>."
• Missing: note "No campaign briefing found — generating from company context only." and continue.

Step 3: Infer and confirm email type, audience, and goal

This is the content-production-specific step for email.

1. Use the loaded audience context and campaign brief (if any) to infer:

   • Email type: marketing (campaign/newsletter/nurture) or sales (cold/warm outreach)
   • Audience: B2B or B2C
   • Content goal: thought leadership, awareness, lead generation, nurturing, conversion, or retention/win-back
   • Funnel stage: TOFU, MOFU, or BOFU
   • Audience expertise level: novice, familiar, or expert

   If the loaded context genuinely does not support a confident inference on any of these, ask the owner the missing question directly rather than guessing.

2. Present a one-line inference summary to the user, for example: "Type: Sales outreach · Audience: B2B (mid-market) · Goal: lead generation · Stage: MOFU · Expertise: familiar"

3. Ask the user to confirm or correct before generating. Do not silently apply assumptions.

4. Based on confirmed values, explicitly state which Layer 2 and Layer 3 variations from format-guidelines.md are being applied and why (e.g., "Applying the B2B sales-outreach variant — staged CTA, account-specific social proof, no scarcity framing — and the MOFU funnel profile — comparative proof, medium CTA directness").

Step 4: Ask for the content topic (if not provided)

If the owner has not specified a topic, offer, or key message, ask:

"What should this email be about? You can describe the offer, key message, or paste raw notes and I'll turn them into a draft."

Wait for the answer, then proceed.

Step 5: Select storytelling framework

Read ../_shared/storytelling-frameworks.md and follow its selection process. For marketing email, AIDA (Attention – Interest – Desire – Action) or the Inverted Pyramid are typically strong fits. For sales outreach, a tighter spine usually works better: reason for contact → relevant problem or goal → proof → low-friction next step. Apply the chosen framework as the structural spine.

Step 6: Select persuasion principles

Read ../_shared/persuasion-principles.md and follow its selection process. Pick 1–3 principles that fit the confirmed email type, B2B/B2C audience, and funnel stage — per format-guidelines.md, avoid scarcity/loss framing in B2B contexts unless the deadline is genuinely real, and prefer account-specific social proof over generic claims in sales outreach. Note the choices in working notes.

Step 7: Generate the email

Produce a complete email that:

• Includes a subject line and preheader (mobile-safe length, no spam-trigger language)
• Opens with a hook that states relevance immediately — no pleasantries or self-introduction first
• Applies the confirmed email-type, B2B/B2C, goal, funnel-stage, and expertise-level variant from format-guidelines.md
• Uses the framework chosen in Step 5 as its structural spine
• Weaves in the persuasion principles chosen in Step 6
• Has exactly one primary CTA, phrased as a specific, low-friction action
• Reflects tone and vocabulary from loaded brand voice context
• Ends with a brief, in-voice sign-off (thanks + sender name) — do not draft a full signature block, legal footer, Impressum, or unsubscribe link

Legal footers, Impressum, and unsubscribe links are fixed parts of the sending template (email marketing tool, CRM "send email" feature, or email client) rather than per-email content. Append this fixed note after the draft, not a [TODO: ...] placeholder:

"Footer/legal compliance (Impressum, unsubscribe link, required disclosures) is not included above — that lives in your sending tool's template. Confirm it meets current legal requirements and your company's policy before sending."

Internally verify against the quality checklist in format-guidelines.md before presenting the output.

Present the output in a clearly delimited block:

─────────────────────────────────────────────
Email draft — <email type>
─────────────────────────────────────────────
Subject: <subject line>
Preheader: <preheader text>

<email body>
─────────────────────────────────────────────
Word count: <N> | Subject length: <N> characters

If output is degraded (a required context need is uncovered), prepend:

⚠ DEGRADED OUTPUT — generated without: <list of missing needs>

Step 8: Feedback

This step has two phases:

Auto-store phase. Before asking the user for feedback, review the run for qualifying observations. For each, append one tagged line to .claude/learnings.md (create with standard header if missing), tagged [cc-content:marketing-email]. Qualifies: content preferences or constraints not already in any loaded context file or CLAUDE.md; corrections the user made; project-specific facts that would change future output; accepted/rejected suggestions deviating from best practices. Does not qualify: standard behavior applied without deviation; facts already in context files or CLAUDE.md; anything derivable by re-reading context files; facts semantically equivalent to any existing .claude/learnings.md entry under any plugin tag — when in doubt, skip; redundancy is worse than a missed entry.

Check for the file before appending:

ls .claude/learnings.md 2>/dev/null && echo "exists" || echo "missing"

Standard header when creating the file:

# Learnings

Corrections and feedback collected during content sessions.
Entries are tagged by skill and dated.

---

Explicit feedback. Ask whether the output met expectations. If the user provides a correction, append it as a tagged entry using the same criteria. Confirm total entries written across both phases: "✓ N learning(s) saved to .claude/learnings.md." If the user confirms quality or skips: if any entries were auto-stored, confirm "✓ N learning(s) auto-saved to .claude/learnings.md." Then deliver a closing line — Paste into your email marketing tool. — and exit.