antislop-copywriting

Rewrites marketing copy and prose to remove AI-writing patterns and fabricated claims.

Updated Aug 1, 2026
One-click install
npx skills add https://github.com/dxiiren/project-skeleton --skill antislop-copywriting-dxiiren
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: antislop-copywriting
Source: https://github.com/dxiiren/project-skeleton/tree/main/.claude/skills-optional/anti-slop/antislop-copywriting
Command: npx skills add https://github.com/dxiiren/project-skeleton --skill antislop-copywriting-dxiiren

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI-generated copy is easy to spot: buzzwords, fabricated statistics, em dashes, forced rule-of-three rhythms, and generic upbeat conclusions make prose read as machine-written and erode reader trust. This Skill provides a systematic pattern catalog and audit loop to detect and rewrite those tells while preserving the author's voice. ## Core Features & Use Cases - Pattern Catalog with Before/After Fixes: Covers tone and voice (empty AI vocabulary, significance inflation, weasel attributions, chatbot closers), rhythm and structure (rule of three, staccato drama, false ranges), honesty (fabricated specifics, speculative gap-filling), and hygiene (em dashes, boldface overuse, inline-header lists), each citing the governing core rule. - Anti-Over-Correction Guardrails: Lists what NOT to flag (perfect grammar, formal vocabulary, isolated em dashes) and signs of genuine human writing to preserve, plus optional voice calibration from a user-provided writing sample. - Draft-Audit-Final Loop and Checklist: A three-step review loop plus a copywriting checklist that runs alongside the core antislop Delivery Gate before any copy is delivered. - Use Case: Given a landing page draft full of phrases like "unlock seamless collaboration" and an invented "trusted by 10,000+ teams" claim, rewrite it into specific, evidenced prose with a real voice and no AI tells. ## Quick Start Ask the AI to load the antislop core and this copywriting skill, then rewrite the attached landing-page draft to remove AI-writing patterns without inventing any facts.

Frequently Asked Questions about antislop-copywriting

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I make AI-generated copy sound human?

Run the draft against the pattern catalog: replace empty vocabulary like "unlock" and "seamless" with specific statements, remove fabricated claims, fix rhythm tells like forced rule-of-three, and cut em dashes. Then read it aloud and verify every fact traces to the source or the user.

What are the most reliable signs of AI-written text?

Look for clusters, not isolated tells: em dashes plus rule-of-three lists plus buzzwords like "vibrant tapestry" plus a generic upbeat conclusion. A single em dash or one short emphatic sentence is not evidence on its own.

Should I remove every em dash and formal word when editing copy?

No. Em dashes are banned in new output per core rule R-02, but if the user provides a writing sample that uses them, match the sample's frequency instead. Formal vocabulary and polished grammar are not AI tells and should be preserved.

Can the rewrite add statistics or testimonials to strengthen weak copy?

No. Rules R-17, R-18, R-36, and R-38 forbid inventing numbers, names, dates, quotes, or citations. If a sentence needs real detail to work, ask the user for it or write the plain version without it.

When should this copywriting skill be loaded?

Load it together with the antislop core whenever the task writes or edits prose meant for people to read: marketing copy, product copy, landing pages, headlines, and CTAs. The core holds the mechanism and hard bans; this skill adds copy-specific depth.