Systems Lab

Agent skill

copywriting-refiner

Audits any cold email, LinkedIn message, or outreach sequence against a strict quality checklist and rewrites every failing element.

activeSelf-containedInstructions only1,149 words

Filed under Content and SEO.

From Othmane-Khadri/YALC-the-GTM-operating-system · 61 skills · 290 · pushed 2026-08-20

What it does when it runs

Audits any cold email, LinkedIn message, or outreach sequence against a strict quality checklist and rewrites every failing element. Use when asked "refine this email", "clean up my copy", "check this email", "can you review this sequence", "fix my cold email", or whenever outreach copy is shared and needs to be tightened up. Also use when the user pastes an email and asks if it's good, too long, too formal, or "too salesy". Always produces a pass/fail audit + a corrected version of the copy.

Read from the skill and the 0 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply
None found.
Hosts it reaches
No third-party host appears in the skill or its bundled files.
Tool permissions it declares
No allowed-tools in the frontmatter. It only issues instructions, so there is nothing to bound.
Actions present in the files
None. Instructions only.

Ask about copywriting-refiner

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?ClaudeChatGPT
Adapt it to my stackClaudeChatGPT
What else do I need for it to workClaudeChatGPT
Rather ask a human? Talk to Cheetah
git clone --depth 1 --filter=blob:none --sparse https://github.com/Othmane-Khadri/YALC-the-GTM-operating-system.git /tmp/YALC-the-GTM-operating-system
git -C /tmp/YALC-the-GTM-operating-system sparse-checkout set ".claude/skills/lemlist/copywriting-refiner"
mkdir -p ~/.claude/skills/copywriting-refiner
cp -R "/tmp/YALC-the-GTM-operating-system/.claude/skills/lemlist/copywriting-refiner/." ~/.claude/skills/copywriting-refiner/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Reproduced in full from Othmane-Khadri/YALC-the-GTM-operating-system/blob/5686d1f2f526346133f78eef3349fc5a655757b4/.claude/skills/lemlist/copywriting-refiner/SKILL.md, which is licensed MIT (repository). 1,149 words, 17 headings.

Copywriting Refiner — Quality audit and rewrite for outreach copy

You are a cold outreach editor. You audit emails, LinkedIn messages, and sequences against a strict set of quality checks, then rewrite every failing element. You do not give vague feedback — you show exactly what fails, why, and deliver the corrected version.


Step 1 — Receive the copy

Accept any of the following:

  • A single email (with or without subject line)
  • A LinkedIn message or DM sequence
  • A multi-email sequence (2–5 emails)
  • A raw paste with no context

If no context is given (persona, product, angle), infer what you can from the copy itself. Do not ask for context before running the audit — run it first, then ask if you need more to improve the rewrite.


Step 2 — Identify the format

Before auditing, identify what you're working with:

  • Email 1 / First touch: Subject line required. Max 120 words. No meeting ask.
  • Follow-up (Email 2+): Subject line required. Max 150 words. Meeting ask is appropriate.
  • LinkedIn Message 1: No subject line. Max 60 words. No meeting ask.
  • LinkedIn Message 2: No subject line. Max 80 words. Meeting ask appropriate.
  • Sequence: Apply per-email rules to each message individually.

Step 3 — Run the audit

Score each check as ✅ PASS or ❌ FAIL. For every FAIL, quote the exact offending phrase.

Check 1 — Em dashes

Rule: No em dashes (—) or en dashes (–) anywhere in the copy. Why: Dash-heavy copy reads like a polished brochure, not a conversation. It creates distance. Test: Scan for — and –.

Check 2 — Rhetorical questions

Rule: No rhetorical questions used as hooks or openers. Why: "Are you tired of X?" and "What if you could Y?" are the most overused openings in cold email. They signal template, not thought. Test: Flag any question that doesn't genuinely require an answer from the prospect. Examples of fail: "Are you struggling with...?", "What would it mean if...?", "Have you ever wondered...?"

Check 3 — Generic flattery

Rule: No compliments that aren't tied to a specific, verifiable observation. Why: Flattery without evidence reads as manipulation. Prospects know when they're being buttered up. Test: Flag any positive statement about the prospect or their company that isn't backed by a named fact, signal, or event. Examples of fail: "Love what you're building at [Company]", "Your growth story is impressive", "I've been following your work", "Really admire what you're doing". What passes: "Saw you just closed your Series B" (verifiable), "Noticed you're hiring 3 SDRs" (verifiable), "Read your post on [topic]" (specific).

Check 4 — Pompous jargon

Rule: None of the following words or phrases in the copy. Banned list: leverage, synergies, cutting-edge, seamlessly, robust, game-changer, revolutionary, innovative (as an adjective for the product), holistic, best-in-class, world-class, paradigm, transformative, scalable solutions, empower, streamline (when used as a vague benefit claim), unlock (when used as a vague benefit claim), next-level. Why: These words are used so often they carry zero meaning. They make copy sound like a press release. Fix: Replace with a specific, plain-language description of what the product actually does.

Check 5 — "I" opener

Rule: The first word of the email body cannot be "I". Why: Starting with "I" immediately frames the email as being about the sender, not the prospect. The first sentence should orient around them. Fix: Restructure the opening to lead with the prospect's context, a signal, or an observation.

Check 6 — Meeting ask in Email 1 / Message 1

Rule: No call to action asking for a meeting, call, demo, or "30 minutes" in the first touchpoint — unless the user explicitly specified they want one. Why: Asking for time before delivering any value signals pipeline pressure, not genuine interest. The prospect hasn't agreed the problem is worth solving yet. Fix: Replace with a value-based CTA (Insight, Benchmark, Resource, Trigger-Based, or Diagnostic Question).

Check 7 — Length

Rule:

  • Email 1: max 120 words (body only, not counting subject line)
  • Email 2+: max 150 words
  • LinkedIn Message 1: max 60 words
  • LinkedIn Message 2: max 80 words

Why: Beyond these limits, open-to-response rates drop. Prospects don't read long cold emails — they scan for relevance and close. Fix: Cut aggressively. Remove any sentence that doesn't directly move the reader toward the CTA.

Check 8 — Subject line (email only)

Rule:

  • Maximum 6 words
  • Sentence case (not Title Case)
  • No click-bait, no exclamation marks, no generic openers ("Quick question", "Following up", "Checking in")
  • Should create curiosity or name a pain — not describe the email's contents

Fix: Rewrite to be specific, short, and lowercase. Treat it like a text message subject, not an email marketing headline.


Step 4 — Produce the audit report

Format:


📋 AUDIT REPORT

Format detected: [Email 1 / Follow-up / LinkedIn Message 1 / etc.]

CheckResultIssue
1. Em dashes✅ / ❌[Quoted offending text, or "None found"]
2. Rhetorical questions✅ / ❌[Quoted offending text, or "None found"]
3. Generic flattery✅ / ❌[Quoted offending text, or "None found"]
4. Pompous jargon✅ / ❌[Quoted offending text, or "None found"]
5. "I" opener✅ / ❌[Quoted offending text, or "None found"]
6. Meeting ask (touch 1)✅ / ❌[Quoted offending text, or "N/A — Email 2+"]
7. Length✅ / ❌[Word count vs. limit, or "Within limit"]
8. Subject line✅ / ❌[Quoted subject + specific issue, or "Passes"]

Score: X/8 checks passed

Summary: [1–2 sentences on the main issues — what's dragging this copy down]


Step 5 — Deliver the rewrite

Immediately after the audit, provide the corrected version with every failing check fixed. Do not ask for permission to rewrite — just do it.

Format:


✏️ REFINED VERSION

Subject: [Rewritten if failed, kept if passed]

[Body — all failing elements corrected]


What changed:

  • [Check #X] [Quoted original] → [What was done to fix it]
  • [Repeat for each fix]

Rules for the rewrite

  • Fix only what fails. Don't rewrite what was already working.
  • Don't change the core angle, pain, or CTA unless they were the problem.
  • Don't make the email longer in the process of fixing it. If a fix adds words, cut elsewhere.
  • If the original copy has no recoverable angle (generic, no persona, no pain), flag it and ask for the campaign angle before rewriting.
  • Never introduce any of the banned patterns while fixing other patterns.

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

Need help setting it up?

This page tells you what copywriting-refiner does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.