Copywriting logo

Copywriting

Community
nahid-sparktales
copywriting

Write persuasive copy that is also true — landing pages, product messaging, emails, ads, launch announcements — by fixing the reader, the objection, the offer and the voice, then auditing every claim against evidence before it ships. Use when copy needs writing or rewriting, when a page says nothing specific, or when a draft's claims need checking before publication. Not for interface strings inside a product, not for documentation, and never authorization to publish or send what it produces.

Overview

Publishernahid-sparktales
Repositoryagent-dispatcher
Skill namecopywriting
Stars
58
Forks
5
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by nahid-sparktales on GitHub. Read the source before you install it.

Installation

Install the Copywriting AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/nahid-sparktales/agent-dispatcher/tree/main/skills/knowledge/copywriting
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/nahid-sparktales/agent-dispatcher.git /tmp/agent-dispatcher
mkdir -p .claude/skills
cp -r /tmp/agent-dispatcher/skills/knowledge/copywriting .claude/skills/copywriting
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Copywriting in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Copywriting on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Copywriting is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Copywriting

Persuasive writing fails in two directions. Vague copy says nothing a competitor could not say. Overstated copy says something the product cannot do. The work is finding the sentence that is both specific and true — and refusing to close the gap with an adjective.

Keep the verbs apart: drafted is written; reviewed is read by someone else; approved is a person accountable for the claims signing off; published is live. This skill gets you to drafted. The last two are not yours to perform.

When this fires

Copy is being written or rewritten for an audience outside the product — a page, an email, an ad, an announcement, a product description — or an existing draft's claims need auditing. It does not fire for buttons, errors and empty states inside the interface, or for documentation.

Procedure

  1. Reconstruct the brief if you were not given one: who the reader is, where they are standing (what they clicked to get here, what they already believe), the one action you want next, and what happens if they do nothing. Infer the low-risk parts; ask about the parts that change the copy.
  2. Gather evidence before writing a word. Approved claims, real numbers with their source and date, actual product behaviour, support tickets, review and sales-call language. Copy is assembled from this pile. When the pile is empty, the honest output is specific-and-modest, not vague-and-grand.
  3. Name the one objection that decides it — too expensive, too slow to switch, I do not believe you, I already have something. Write it down. Every section either advances the promise or answers that objection; anything doing neither is cut.
  4. State the promise in one plain sentence before drafting, in the reader's words rather than the company's. If it could sit unchanged on a competitor's page, it is not yet the promise.
  5. Establish voice from existing material, not from a default marketing register: read published pages, the product's own interface, the founder's writing. Match register, sentence length and vocabulary. Preserve the user's voice when rewriting rather than replacing it.
  6. Structure for the channel. A page: one proposition, proof beneath it, one action. An email: one reason it arrived, one thing to do. An ad: one idea. Subject lines, headlines and first lines carry most of the weight — write several and pick, rather than polishing the first.
  7. Draft with the specific detail instead of the intensifier. The number, the named integration, the actual time saved, the real constraint. "Blazing fast" is a request for the reader's trust; "renders 10k rows without pagination" spends evidence instead.
  8. Audit every claim, line by line, and sort each into: verifiable (name the source), opinion (marked as a view, not a fact), puffery (harmless enthusiasm), or unsupported (cut or downgrade). Pay particular attention to numbers and percentages, comparatives naming or implying a competitor, superlatives and firsts, guarantees, refund and uptime promises, and the words secure, private, compliant and certified — each of which asserts something auditable.
  9. Never manufacture proof. No invented testimonials, customer names, logos, case studies, statistics, awards or press quotes, not even as placeholders that could be mistaken for real ones. A placeholder handed to you in a brief is a placeholder, not a customer quote — label it and ask.
  10. Route regulated claims to a human. Health, medical, financial and earnings, legal, safety, environmental and accessibility claims, and anything naming a certification or a competitor: propose the wording, mark it as needing approval, and do not treat your own confidence as clearance.
  11. Cut, then read it aloud. Remove the throat-clearing opener, the repeated proposition, the stacked adjectives, the jargon the reader would not use. What is left should sound like a person who knows the product, not like a category.
  12. Deliver as a draft. Publishing, posting, sending to a list, or updating a live page is an outward-facing action performed by the user or on their explicit per-item say-so: stop and ask. Hand over the copy, the claims list, and what still needs approval.

Checklist

  • Reader, moment, desired action and cost of inaction written down
  • Evidence gathered before drafting; every proof point traceable to a source
  • The deciding objection named and answered somewhere in the copy
  • Promise stated in one sentence a competitor could not reuse
  • Voice taken from existing material, not a default register
  • Structure fits the channel; one action, not three
  • Specifics carry the argument; intensifiers are not doing load-bearing work
  • Every claim sorted: verifiable, opinion, puffery, or cut
  • No invented testimonial, logo, statistic, award or case study
  • Regulated and competitor claims flagged for approval, not self-cleared
  • Delivered as a draft; nothing published or sent

Failure handling

  • The product cannot do what the brief promises — say so and stop. Rewriting around it produces copy that converts into refunds and support load. Offer the true version of the claim.
  • No evidence exists for the central proof point — do not smooth it into a superlative. Write the smaller true claim and name the missing evidence as a blocker.
  • The brief supplies a testimonial or logo with no source — treat it as unverified and ask before using it. Placeholder proof shipped as real proof is the trap here.
  • You are asked to match a competitor's stronger claim — their wording is not evidence for yours. Compare against what you can document.
  • The real problem is the offer, not the copy — say it. Copy that has to apologize for the product is a positioning finding, and writing around it hides the finding.
  • A claim needs a decision only the client can make — present the options with the risk of each; do not pick the bolder one because it reads better.

Evidence to report

The brief as you reconstructed it, marked where you inferred rather than were told; the draft itself as the main deliverable; a claims list pairing each factual assertion with its source; the claims you downgraded or cut and why; what still needs human approval before it can run; and the explicit note that nothing was published or sent. "Copy delivered" with no claims list is the version of this work that gets somebody in trouble.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Copywriting AI skill do?

Write persuasive copy that is also true — landing pages, product messaging, emails, ads, launch announcements — by fixing the reader, the objection, the offer and the voice, then auditing every claim against evidence before it ships. Use when copy needs writing or rewriting, when a page says nothing specific, or when a draft's claims need checking before publication. Not for interface strings inside a product, not for documentation, and never authorization to publish or send what it produces.

Why use Copywriting on TypingMind?

Because you install it once and use it with any model. Copywriting is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Copywriting in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/nahid-sparktales/agent-dispatcher/tree/main/skills/knowledge/copywriting. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Copywriting?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Copywriting?

As many as you like. As long as a model supports skills, you can use Copywriting with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Copywriting AI skill free?

Yes. It is published on GitHub by nahid-sparktales under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇