agentleFS
busyeugenebusyeugene/content-marketing-skills/email-copy/SKILL.md

email-copy

Skill1 starsChanged 7 months agoMITView on GitHub

Use when the user wants to draft marketing or lifecycle email copy — campaign emails, nurture sequences, product announcements, cold outbound, or newsletters. Produces subject-line variants and a full body with CTA.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 7 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use AskUserQuestion for email type, length mode, and sequence Y/N.

    line 29
  2. Very specific:

    Read editorial-guidelines.md or fall back to shared voice reference.

    line 35
  3. Distinctive:

    Read any context asset the user pointed to.

    line 36
  4. Distinctive:

    Before writing, articulate in one sentence each:

    line 40
  5. Distinctive:

    If any of the three is unclear, ask the user before drafting.

    line 45
  6. Distinctive:

    No walls of text: paragraphs of 1–3 sentences.

    line 90
  7. Distinctive:

    If the user asked for a sequence:

    line 98
  8. Distinctive:

    No editorial guidelines: use shared voice reference.

    line 156
  9. Distinctive:

    Insufficient context: ask targeted questions before drafting.

    line 157

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Email Copy

Draft email copy that matches the brand's editorial voice. Covers one-off campaigns, lifecycle sequences, newsletters, announcements, and cold outbound. Output is local-only — this skill does not send.

Setup

No keys required.

Inputs

  1. Email type — campaign / nurture sequence / product announcement / cold outbound / newsletter issue / transactional follow-up.
  2. Goal — what does a successful email cause the recipient to do? (click, reply, book, read, reactivate).
  3. Audience segment — who is receiving it? Inherit from editorial-guidelines.md audience if available; otherwise ask.
  4. Stage in relationship — cold, warm lead, trial user, paying customer, churned.
  5. Primary CTA — exact action and destination URL or reply prompt.
  6. Context assets — optional: blog post or landing page the email is tied to; any specific proof points, numbers, or quotes.
  7. Length mode — short (≤120 words), medium (120–250), long (250–500).
  8. Number of subject-line variants — default 5.
  9. Sequence mode — if the user wants more than one email, how many, what cadence, and what's the goal of each step.

Use AskUserQuestion for email type, length mode, and sequence Y/N.

Process

1. Load voice

  • Read editorial-guidelines.md or fall back to shared voice reference.
  • Read any context asset the user pointed to.

2. Map the message

Before writing, articulate in one sentence each:

  • The one thing the email makes the reader feel
  • The one thing the email makes the reader think
  • The one thing the email makes the reader do

If any of the three is unclear, ask the user before drafting.

3. Draft subject lines

Produce N variants across these archetypes (pick the ones that fit the email type):

  • Specific number — "How Acme cut ramp time to 11 days"
  • Curiosity gap — "The reason your SQL→SAL conversion is lying to you"
  • Direct benefit — "Book 2× more demos this quarter"
  • Pattern interrupt — "Bad news about your pipeline dashboard"
  • Personal/low-fi — "Quick question, {name}"
  • Announcement — "{Feature} is live"
  • Contrarian — "Stop A/B testing your cold emails"

Each variant: ≤50 characters. Include one preheader (≤90 chars) per subject. Never repeat phrasing across variants.

4. Draft the body

Structure depends on type:

Campaign / nurture / cold outbound:

  1. Opening line — specific to the recipient or the moment; never "Hope this email finds you well".
  2. The reason you're writing — one sentence, no preamble.
  3. The value or insight — one concrete thing (number, framework, observation). Not a pitch.
  4. The CTA — one action, one link. Phrased as a low-friction next step, not a demand.
  5. Signature — name, role, (optional) one-liner context.

Product announcement:

  1. What's new (one line).
  2. Who it's for.
  3. Why we built it (the specific problem).
  4. What to do next (demo, docs, feature toggle).

Newsletter issue:

  1. One-paragraph editor's note (human, specific, opinionated).
  2. 3–5 curated sections, each 40–80 words, each with a link.
  3. A closing "what I'm thinking about this week" prompt that invites reply.

Transactional follow-up:

  1. Acknowledge the event (trial started, demo booked, payment processed).
  2. The one thing they should do in the next 24 hours.
  3. Where to get help.

5. Constraints

  • Mobile-first: first 60 characters of subject + first sentence of body must carry the message.
  • No walls of text: paragraphs of 1–3 sentences.
  • Banned words from editorial guidelines are blocked.
  • First-person singular is OK for cold outbound and newsletter; first-person plural for campaigns and announcements — unless guidelines say otherwise.
  • Zero emojis unless the guidelines allow them.
  • Exactly one CTA. If the user insists on two, demote the second to a PS.

6. Sequences

If the user asked for a sequence:

  • Draft each email as a separate block in the same output file.
  • Each email references the previous one implicitly (shared context) but must be readable standalone (the reader may have missed step 1).
  • Cadence recommendation included at the top of the file.
  • The last email in a sequence includes a graceful opt-out line.

7. Write the file

Write to emails/<name>.md:

# Email — {name}

**Type:** {type}
**Audience:** {segment}
**Goal:** {goal}
**Primary CTA:** {action → destination}
**Length mode:** {short/medium/long}

## Subject line variants

1. **Subject:** {…} — **Preheader:** {…}
2. …

## Body

{body}

---

## Sequence (if applicable)

### Step 2 — send +{N} days

**Subject:** {…}
**Preheader:** {…}

{body}

{…}

## Recommended cadence

- Step 1: day 0
- Step 2: day +{N}
- Step 3: day +{M}

## Rationale

{2–4 bullets on why this approach for this audience}

8. Print summary

One-line summary: path, word count, number of subject variants, sequence length.

Fallbacks

  • No editorial guidelines: use shared voice reference.
  • Insufficient context: ask targeted questions before drafting; never guess the audience or CTA.

Verification

  1. File exists under emails/.
  2. ≥5 subject lines, each ≤50 chars.
  3. Each subject has a preheader ≤90 chars.
  4. Body matches the length mode.
  5. Exactly one primary CTA.
  6. No banned words.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.