anti-ai-slop-writing — Claude Code skill
This skill applies writing rules that reduce common AI-style patterns. Use it when drafting or rewriting emails, articles, README files, posts, and similar copy.
npx skills add jalaalrd/anti-ai-slop-writing --skill anti-ai-slop-writing -a claude-codeWhat anti-ai-slop-writing does
anti-ai-slop-writing activates for writing and rewriting requests. It steers the agent toward direct wording, concrete details, connected sentences, and a voice suited to the format. The guidance covers emails, posts, reports, messages, descriptions, and README files.
Before drafting, the skill loads a blocked-language list from references/banned-words.md. It also asks for varied sentence lengths and paragraph shapes, active phrasing, and factual restraint. Its checks cover punctuation too, including limits for em dashes, exclamation marks, and ellipses.
The rules are deliberately strict, so familiar corporate phrasing may need to go. That is the trade-off for text that does not fall into repeated generated-text habits. The skill also asks for a clear position when the evidence supports one.
What it helps with
- rewrite an email without stock AI phrasing
- draft a README with a natural voice
- break repetitive sentence patterns in a post
- remove blocked wording from a draft
- keep claims free of invented figures or quotes
Example prompts for the agent
- Rewrite this email in a natural voice without AI-style phrasing.
- Draft a project README and follow anti-ai-slop-writing.
- Make this post sound less formulaic while keeping the facts.
How to install anti-ai-slop-writing
1Skills CLI (Vercel)
npx skills add jalaalrd/anti-ai-slop-writing --skill anti-ai-slop-writing -a claude-code2Manually: put the files in place
~/.claude/skills/anti-ai-slop-writing/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/anti-ai-slop-writing/
1Skills CLI (Vercel)
npx skills add jalaalrd/anti-ai-slop-writing --skill anti-ai-slop-writing -a codex2Manually: put the files in place
.agents/skills/anti-ai-slop-writing/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add jalaalrd/anti-ai-slop-writing --skill anti-ai-slop-writing -a cursor2Manually: put the files in place
.cursor/skills/anti-ai-slop-writing/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install jalaalrd/anti-ai-slop-writing anti-ai-slop-writing2Manually: put the files in place
.github/skills/anti-ai-slop-writing/SKILL.md1Skills CLI (Vercel)
npx skills add jalaalrd/anti-ai-slop-writing --skill anti-ai-slop-writing -a gemini-cli2Platform CLI
gemini skills install https://github.com/jalaalrd/anti-ai-slop-writing --path skills/anti-ai-slop-writingManually: put the files in place
.devin/skills/anti-ai-slop-writing/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add jalaalrd/anti-ai-slop-writing --skill anti-ai-slop-writing -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about anti-ai-slop-writing
How do I install anti-ai-slop-writing in Claude Code?
Run npx skills add jalaalrd/anti-ai-slop-writing --skill anti-ai-slop-writing -a claude-code.
What does anti-ai-slop-writing do?
The skill applies constraints for drafting and rewriting text to avoid common generated-writing patterns. It covers wording, structure, punctuation, factual accuracy, and voice.
Which writing rules does the skill apply?
It loads blocked terms from references/banned-words.md, asks for varied sentence length, and discourages chains of short statements. It sets a limit of one em dash per 500 words.
