Individual skill
anti-slop
Strip 85 stylometric, syntactic, cross-linguistic, and marketing-hype AI writing tells across any language. Use when drafting, editing, or humanizing text alongside tools/slop_scan.py.
skills/anti-slop/SKILL.md
Source commit 953e486421606632752c455b4ea3e8305220f2c0
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- No
- Repository scan complete
- No
- Skill scan complete
- No
- Observed behavior
- Not checked yet
- Last checked
- Nov 8, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 19
- Repository last updated
- Nov 3, 2026
- License
- MIT
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- agent Repository topic
- agent-skill Repository topic
- agent-skills Repository topic
- chatgpt Repository topic
- claude Repository topic
- claude-ai Repository topic
- claude-skill Repository topic
- claude-skills Repository topic
- gemini Repository topic
- plugin Repository topic
- writing Derived category
- marketing Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 22,444 bytes
Read excerpt
# Universal Anti-Slop & Anti-Hype (85 Deep Stylometric & Cognitive Rules)
Why can experienced readers still "smell" AI writing or cheap marketing even after you delete obvious words like `delve`, `tapestry`, and `seamless`? Because Large Language Models have a deeper **cognitive and syntactic signature**: uniform information entropy per sentence, see-saw concession clauses, polite professorial neutrality, abstract noun chains, symmetrical paragraph geometry, and forced trend-jacking.
Apply all 85 rules across **any human language** (English, Vietnamese, Spanish, French, German, Japanese, Chinese, Korean) to strip both surface cliches and deep structural AI tells.
---
## Dimension 1: Information Entropy & Discourse Geometry (Rules 1-10)
1. **Kill Uniform Information Entropy** because LLMs distribute information evenly so every sentence carries roughly the same lukewarm weight. Human writers compress high-surprise facts into dense bursts, followed by ultra-short punchlines.
2. **Ban the 3-Sentence "Sandwich Paragraph"** where LLMs follow the exact formula `[Sentence 1: Abstract claim] -> [Sentence 2: Dual/triple example] -> [Sentence 3: Moralizing summary tail]`. Break this mold: let a paragraph be a single 6-word verdict, or a 5-sentence causal narrative.
3. **Vary Paragraph Visual Density** by never writing four consecutive paragraphs of 3-4 lines each. Alternate short and long blocks so the page looks human-typeset before a single word is read.
4. **Delete Throat-Clearing Macro Openers** that announce the world is changing or a topic is important (`"In today's..."`, `"Trong bối cảnh..."`, `"En el mundo actual..."`). Start with a concrete actor, number, or event in Sentence 1.
5. **Delete Summary Echo Closers** starting with `"In conclusion,"`, `"Ultimately,"`, `"All in all,"`, or `"Tóm lại,"` that repeat the thesis. End at peak utility or on a concrete image.