Skip to content
SkillModel-led4,027 tokens in use

copy-editing

Edits existing marketing copy in seven focused sweeps, scores high-stakes pages with an expert panel and strips AI tells while keeping the author's message.

61,711 stars on the repository MIT Licensecoreyhaines31/marketingskills · skills/copy-editing
npx skills add coreyhaines31/marketingskills --skill copy-editing
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing copy-editing costs

103tokens idle
4,027tokens in use
8,314tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter12/15
  • Structure19/25
  • Examples6/15
  • Output15/15
  • Size15/15
  • Resources15/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity100
  • Maintenance100
  • Community76
  • Maturity32
  • Momentum40

copy-editing improves copy that already exists. The agent first picks a mode: rebuild from the facts, restore a real voice, or make a point edit to one line. Full edits run seven sweeps in order: clarity, voice and tone, so what, prove it, specificity, heightened emotion and zero risk, looping back after each.

It also runs an AI-tell check on every edit and offers a content refresh routine for pages whose stats or examples have gone stale.

What You Get

  • Edits from seven single-purpose sweeps, each with a clear reason
  • Point edits that return only the changed lines with a one-line reason
  • A multi-persona expert panel score for launch, pricing and landing copy
  • A quick-pass list of words to cut and weak words to replace
  • A refresh plan for outdated pages with a refresh or rewrite decision

Common Use Cases

  • Tightening a landing page draft before launch
  • Fixing one headline without touching the rest of the page
  • Removing contrast reveals and stacked negation lists from AI-written copy
  • Refreshing an old post whose traffic is falling

What Makes It Unique

A point edit skips the sweeps and changes only the requested line. The AI-tell check names the shapes that give machine writing away (contrast reveals, negation lists, trailing clauses, colon reveals) and fixes them from the underlying facts. A missing fact is marked as a need the author fills in. An expert panel gate sets a bar of 7 per persona and 8 on average.

Under The Hood

The SKILL.md and three reference files are loaded as context. It ships no scripts and runs no commands. The only file it reads on its own is the product marketing context file in your project. Edits are returned in chat or applied to the copy you point at.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers