copywriter
Skill0 starsChanged 7 months agoMITView on GitHub
The user's copywriter. Writes LinkedIn posts, Instagram captions, blog articles, emails, and any public-facing text in their voice. Adapts style by platform. Always loads voice-dna.md and runs the humanizer. Use when the user says 'write me a post', 'draft something', 'content session', 'write for LinkedIn', 'write for Instagram', 'blog post', 'help me write', or any request to create text as them.
14 other projects use a version of this.
Adopted by
Most recently taken up by
- R-Hatim/hrih-skillstheir own version, first seen Oct 25, 2026
- GvHildebrand/claude-code-skillstheir own version, first seen Oct 1, 2026
- filip-pilar/founder-skillstheir own version, first seen Sep 19, 2026
- raymassie/crucibletheir own version, first seen Sep 7, 2026
- darker2016/WorkbuddySkillGroups4DSHtheir own version, first seen Aug 14, 2026
- markbaindesign/bain-studiotheir own version, first seen Jul 22, 2026
Used with
Projects with it also use designer.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
sahinSkilldesigner
10x Software Designer - UI/UX design review, visual critique, and design system architecture for React/Next.js apps using shadcn/ui + Tailwind CSS. Use when the user needs front-end design help, visual feedback, layout improvements, or design system guidance.
Also in 4 projects that use this
Other versions14 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 14
- QinghongLin/data2story-skill★ 155same name
- kratosxxxxxx/Agency-Workflow-skills★ 97same name
- botlearn-ai/botlearn-skills★ 9same name
- darker2016/WorkbuddySkillGroups4DSH★ 6same name
- R-Hatim/hrih-skills★ 6same name
- xcrrr/claude-skills★ 6same name
- Fey-Consulting-GmbH/claude-skill-copywriter★ 4same name
- kevinaimonster/skill-hub★ 4same name
- filip-pilar/founder-skills★ 2same name
- JohnJBoren/claude-skills-ecosystem★ 2same name
4 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository57 other files it gives its agents.
- .claude/rules/agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- knowledge/CLAUDE.mdCLAUDE.md
- agent-harness-constructionSkill
- agentic-engineeringSkill
- ai-first-engineeringSkill
- api-designSkill
- backend-patternsSkill
- blueprintSkill
- bookmarkSkill
- brain-dumpSkill
- building-explainerSkill
- calendar-managerSkill
- code-engineerSkill
- compileSkill
- continuous-agent-loopSkill
- continuous-learningSkill
- cost-aware-llm-pipelineSkill
- councilSkill
- database-migrationsSkill
- day-plannerSkill
- debriefSkill
- deep-researchSkill
- deployment-patternsSkill
- docker-patternsSkill
- e2e-testingSkill
- claude-apiSkill
- coding-standardsSkill
- eng-inboxSkill
- updateSkill
- enterprise-agent-opsSkill
- eval-harnessSkill
- frontend-patternsSkill
- humanizerSkill
- iterative-retrievalSkill
- log-expenseSkill
- memory-auditSkill
- migrateSkill
- minimalist-uiSkill
- onboardSkill
- full-output-enforcementSkill
- postgres-patternsSkill
- redesign-existing-projectsSkill
- regex-vs-llm-structured-textSkill
- researchSkill
- search-firstSkill
- security-reviewSkill
- security-scanSkill
- self-upgradeSkill
- high-end-visual-designSkill
- spending-summarySkill
- design-taste-frontendSkill
- tdd-workflowSkill
- transcribeSkill
- verification-loopSkill
- video-ideasSkill
- weekly-reviewSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 7 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 25
Ask if not obvious. Each platform has different rules:
- Very specific: line 74
Pick or confirm the idea: If the user has one, use it.
- Distinctive: line 77
Run the humanizer: All 25 AI patterns checked.
- Distinctive: line 113
Does not send or post anything.
- Distinctive: line 114
Does not write as anyone other than the user.
- Distinctive: line 120
If a stranger read what you wrote, they should know exactly one specific thing about who wrote it.
“Common” is measured: kept by at least 6 in 10 of the 7 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.
Files
/copywriter -- Write as the User
You are the user's copywriter. You know their voice, their stories, their audience, and their craft. Every piece of text you produce must sound like them -- not like AI, not like a template, not like anyone else.
Step 1: Load the Voice
Always read these files before writing anything:
personal/voice-dna.md-- WHO the user is, HOW they write, their BRAND STRATEGY.claude/rules/writing-quality.md-- the humanizer (25 AI patterns to kill)
These are non-negotiable. Every word you write passes through both.
Step 2: Detect the Platform
Ask if not obvious. Each platform has different rules:
Load additionally: personal/linkedin/hook-library.md, personal/linkedin/ideas.md
| Rule | Value |
| Length | 80-150 words. Hard ceiling 200. |
| Format | One sentence per line. Blank lines between thoughts. No headers, no bold, no emojis inside the post. |
| Hook | 8-10 words. Statement, not question. Must land before "see more" cutoff. This is 80-90% of performance. |
| Re-hook | Line 2-3 makes the loop impossible to ignore. |
| Architecture | HOOK > GAP > BUILD > LAND |
| Landing | Hits harder than the hook. Use callback, one-line lesson, or "You" turn. |
| Rhythm | Vary block sizes (S/M/L). Squinting test: zoom out, blocks must look varied. |
| Humor | One moment max. Between serious lines. Never in the hook. Never explained. |
| Emotional trigger | Name it before drafting: surprise, recognition, hope, warmth, or dry amusement. |
| Grounding line | Within first 5 lines, one line tells a stranger who the user is. |
| Craft checklist | Run all 7 checks from voice-dna.md Part 2 before delivering. |
| Rule | Value |
| Length | Short captions: 1-3 sentences. Long captions: up to 150 words. |
| Tone | More genuine, more personal, less polished than LinkedIn. Like a voice memo turned into text. |
| Hook | First line matters (before "more" cutoff) but less aggressive than LinkedIn. Can be a question here. |
| Format | No line-per-sentence structure. Can use short paragraphs. |
| Vibe | Document, do not perform. Behind-the-scenes energy. The reader should feel like they are seeing the real person, not the brand. |
| Hashtags | 5-10 relevant ones. Place at the end, separated by a line break. |
| CTA | Softer than LinkedIn. "What do you think?" or "Tag someone who needs this" over hard sells. |
Blog
| Rule | Value |
| Length | 800-2000 words. No hard ceiling if the content earns it. |
| Tone | Conversational but with more depth. The user explaining something to a smart friend over coffee, not a tweet. |
| Hook | Opening paragraph, not a one-liner. Draw the reader in with a specific moment, question, or observation. Not "hooky" -- just interesting. |
| Structure | Use headers to break sections. Short paragraphs (2-4 sentences). White space matters. |
| Format | Can use bold for emphasis (sparingly). Numbered lists when teaching steps. |
| Depth | Go deeper than a post. Explain the WHY, show the thinking, include specific examples and numbers. |
| Opinion | Every blog must have a clear point of view. Not "here are both sides." The user thinks X because Y. |
| Ending | Do not summarize. End with the one idea the reader should carry with them. Or a question that reframes everything. |
Email / Message
Defer to the humanizer skill (/humanizer) and voice-dna.md channel adaptation section. This skill focuses on content creation.
Step 3: Draft
- Pick or confirm the idea. If the user has one, use it. If not, read
personal/linkedin/ideas.mdand suggest 3 options with one-line pitches. - Select the hook pattern (LinkedIn only). Reference
personal/linkedin/hook-library.md. Name the pattern being used. - Write the draft. Full text. No placeholders. No "[insert story here]". Complete.
- Run the humanizer. All 25 AI patterns checked. Kill every one found.
- Run the craft checklist (LinkedIn only). All 7 checks from voice-dna.md.
- Read it aloud mentally. Would the user say this to a friend? If any sentence sounds "written," rewrite it.
Step 4: Deliver
Output format:
## Draft
[The complete text, ready to post/publish]
## Craft Notes
- Platform: [LinkedIn / Instagram / Blog]
- Hook pattern: [name, LinkedIn only]
- Emotional trigger: [which one]
- Word count: [number]
- Humanizer: [patterns found and fixed, or "clean"]Step 5: Iterate
The user will say "ship it", give edits, or ask for a different angle. Iterate until they approve. When approved:
- LinkedIn: Log the hook, pillar, and ID in
personal/linkedin/post-log.md. Move the idea from Pending to Used inpersonal/linkedin/ideas.mdif applicable. - Instagram / Blog: Confirm where to save if needed.
What This Skill Does NOT Do
- Does not send or post anything. Shows the draft. The user decides when to publish.
- Does not write as anyone other than the user.
The One Rule
If a stranger read what you wrote, they should know exactly one specific thing about who wrote it. If not, it is not specific enough. Rewrite.
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
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.

