Claude Code Skills

Output that sounds like you, not a chatbot

Professional Claude Code skills with built-in anti-AI voice rules. Install once, use forever. No more generic AI slop.

Try it free

How it works

1

Download the zip

Each skill is a folder with SKILL.md and supporting files.

2

Copy to Claude

Drop the folder into ~/.claude/skills/ and you're done.

3

Use the command

Type /skill-name in any session. Claude does the rest.

Try it free

Pick a skill, edit the input, and see what it produces. One free generation per visit.

Skills

Each one includes VOICE-RULES.md — our anti-AI detection ruleset.

✍️

LinkedIn Post Pro

Write LinkedIn posts that get engagement, not impressions. Six hook patterns, platform-native formatting, and aggressive anti-AI voice rules.

  • 6 proven hook patterns with rotation
  • 800-1,300 char sweet spot targeting
  • 10+ banned AI writing patterns
  • Mobile-first formatting
📋

Meeting to Actions

Paste messy meeting notes, get structured action items with owners and deadlines. Plus a send-ready follow-up email.

  • Ownership detection from context
  • Deadline conversion (relative to absolute)
  • Unassigned task flagging
  • Send-ready follow-up email
🔍

Competitive Intelligence Brief

Turn any competitor's name or URL into a full strategic intelligence report. Positioning, pricing, strengths, weaknesses, opportunities.

  • 5-section strategic framework
  • Evidence-based, not opinion-based
  • Job posting intelligence
  • Executive TL;DR summary
📅

Content Calendar Generator

Complete content calendar with pre-written hooks, pillar rotation, format variety, and repurposing plan. Any platform.

  • Pre-written hooks for every slot
  • Content pillar rotation
  • 6 platforms supported
  • Repurposing plan included
✉️

Email Tone Rewriter

Rewrite any email in 8 tones: professional, assertive, diplomatic, concise, apologetic, and more. Same message, different impact.

  • 8 distinct tone profiles
  • Tone-specific rewriting techniques
  • Word count tracking
  • Anti-cliche enforcement

Why these aren't just prompts

Anti-AI Voice Rules

Every skill ships with VOICE-RULES.md — a comprehensive banned-phrase list and rhythm guidelines. No "Here's what you need to know." No rules of three. No AI drumroll patterns. Output that reads like a human wrote it.

Proper Skill Format

Not a text file you paste into a chat. These are structured SKILL.md files with YAML frontmatter, supporting files, and examples. Install once, invoke with /command. Claude loads them automatically when relevant.

Self-Checking Output

Each skill includes quality gates. The LinkedIn skill checks if the hook works in isolation. The meeting skill flags unassigned tasks. The email rewriter tracks word count changes. Built-in quality, not afterthought.

Questions

What is a Claude skill?

A skill is a set of instructions that extends what Claude Code can do. You install it by copying a folder to ~/.claude/skills/, then invoke it with a /command. Claude follows the skill's instructions to produce structured, high-quality output for a specific task.

Do I need Claude Code?

Yes. Skills work with Claude Code (Anthropic's CLI tool). You can use the free or paid version. The skills themselves are one-time purchases — no subscription needed.

What's in the download?

A zip file containing: SKILL.md (the main instructions), VOICE-RULES.md (anti-AI detection rules), README.md (documentation), and any supporting files like examples or templates.

Can I modify the skills?

Yes. They're plain markdown files. Edit them however you want. Add your own voice rules, adjust the frameworks, combine multiple skills. You own the files.

What makes these different from free prompts?

Three things: the anti-AI voice rules (most free prompts produce obvious AI output), the proper skill format (install once, use forever via /command), and the self-checking quality gates (output catches its own mistakes before you see them).

Do you offer refunds?

Yes. If a skill doesn't produce the results described, reach out to hello@theskill.dev and we'll work with you to get it right. If we can't resolve it, full refund within 14 days — no questions. Every skill is tested across hundreds of generations before publishing.