agentleFS
forsvn-labsforsvn-labs/meta-skills/skills/write-copy/SKILL.md

write-copy

Skill14 starsChanged 3 months agoMITView on GitHub

Write or rewrite finished product-marketing copy. Use for landing pages, product pages, launch copy, ads, lifecycle email, sales or partnership outreach, announcements, long-form content, messaging variants, headlines, calls to action, or improving copy so it is specific, credible, on-brand, and ready to publish.

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

4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask at most one bundled question when different answers would change the copy materially.

    line 25
  2. Distinctive:

    Prefer concrete nouns, active verbs, observable outcomes, and product details a competitor could not claim unchanged.

    line 30
  3. Distinctive:

    Before delivery, run the competitor-swap, first-screen, evidence-proximity, friction, deletion, and read-aloud tests.

    line 64
  4. Distinctive:

    Never invent claims, quotes, metrics, customers, or consent.

    line 67

“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

Write product-marketing copy

Produce the finished copy first. Use the customer's brand as the brand of record.

Anchor the work

Infer:

  • audience and real moment;
  • product mechanism;
  • credible outcome;
  • proof;
  • main objection;
  • intended action;
  • surface, voice, language, and constraints.

Ask at most one bundled question when different answers would change the copy materially. Otherwise label the assumption and continue.

Write from the product outward

Prefer concrete nouns, active verbs, observable outcomes, and product details a competitor could not claim unchanged. Put proof next to the claim it supports. Match the CTA to current trust.

Remove:

  • “all-in-one,” “seamless,” “revolutionary,” and other category filler without evidence;
  • empty setup, repeated summaries, generic AI cadence, and decorative jargon;
  • fake quotes, invented numbers, vague attribution, or unsupported superlatives;
  • forced urgency, engagement bait, false familiarity, and channel-inappropriate formatting.

Apply the surface:

  • Landing page: audience, costly moment, promise, mechanism, proof, objections, and next step in a scan-friendly sequence.
  • Ads: align hook, promise, proof, visual idea, and landing-page expectation.
  • Lifecycle email: trigger from real behavior, give one useful next step, and define exit or suppression logic when planning a sequence.
  • Outreach: earn relevance with an observed signal, state affiliation, make a small ask, and never pretend familiarity.
  • Launch: explain what changed, who it serves, why now, how it works, and how to try it.
  • Long form: make one defensible argument; use examples and evidence to advance it.

Deliver

Return paste-ready copy in the requested structure. Add variants only when each represents a different strategic bet, and label the bet rather than calling them A/B/C.

Then provide at most:

  • the chosen audience, promise, and mechanism;
  • two or three consequential decisions;
  • factual claims that require verification;
  • one recommended next action.

Before delivery, run the competitor-swap, first-screen, evidence-proximity, friction, deletion, and read-aloud tests. Revise before explaining.

Never invent claims, quotes, metrics, customers, or consent. Keep sending, publishing, and external writes behind explicit approval.

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.