Skip to content
← Tools / Skills
Skill · .md packagevidanov/10 trustMetrics-checked
Writing Craft Skill logo

Writing Craft Skill

Teaches agents to write well: classic copy craft plus an AI-tell checklist.

Visit official site →

About

A writing skill with two halves: craft principles drawn from Ogilvy, Sugarman and Zinsser (word choice, sentence rhythm, loss framing) and a 24-pattern AI-detection checklist covering vocabulary, structure, formatting and rhythm tells. It has write and edit modes, ships as a plugin plus a plain SKILL.md, and installs into 50+ agents via the skills CLI. At 1 star it is still early, and its generative voice leans marketing-craft rather than neutral technical prose. As an anti-AI-tell layer paired with Neuricon's own humanizer rule, it is unusually on-theme.

What it can do

  • Drafting with classic copywriting craft
  • Editing passes with 24-pattern AI-tell removal
  • Voice calibration and tone guidance
  • Grammar-and-clarity-only edit mode

Setup

Run: npx skills add vidanov/writing-craft-skill

Manual: copy SKILL.md to ~/.claude/skills/writing-craft/SKILL.md (or .kiro/skills/writing-craft/)

ChatGPT/Claude Projects: paste chatgpt/PROMPT.md into custom instructions

Start a new task so the agent refreshes its skill inventory

Prerequisites

Any SKILL.md-compatible agent (Claude Code, Cursor, Codex, Kiro); or any chat assistant for the PROMPT.md variant.

Use cases

writingsocial

Works with

Claude CodeCursorCodexkiroChatGPT

Trust score — 5/10

/10 trustMetrics-checked

Scored 5/10 (metrics tier, metrics-v1). Evidence: 1 star, 13 commits, 1 release — early single-author skill with skills.sh manifest.; Install via skills CLI verified; manual single-file and ChatGPT PROMPT.md paths documented.; 24-pattern AI-tell checklist pairs naturally with humanizer-style editing.. License: MIT verified via GitHub license field..

★ 1 GitHub stars

How the trust score is built

Full methodology →
Adoption · stars/downloads, log-scaled vs peers
5.0/10
Maintenance · recency, releases, issue load
6.0/10
Openness · license + docs
7.0/10
Manual review · sanity read + red-flag check
7.0/10
  • · 1 star, 13 commits, 1 release — early single-author skill with skills.sh manifest.
  • · Install via skills CLI verified; manual single-file and ChatGPT PROMPT.md paths documented.
  • · 24-pattern AI-tell checklist pairs naturally with humanizer-style editing.

Sources

Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.