skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
christofferbergj/agent-skills70 installs

unslop

Cut AI tells from Pstack output or from writing submitted for editing.

How do I install this agent skill?

npx skills add https://github.com/christofferbergj/agent-skills --skill unslop
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    This skill consists of a set of stylistic writing rules designed to help the AI remove common "AI-isms" or linguistic patterns often found in model outputs. It contains no executable code, network operations, or sensitive data access, and poses no security risk.

  • Socketpass

    No alerts

  • Snykwarn

    Risk: MEDIUM · 1 issue

What does this agent skill do?

Unslop

Edit text to remove AI patterns. Scan for the patterns below, then rewrite. Preserve meaning and match the intended tone.

Rule numbers are stable ids that other skills cite. A removed rule leaves a gap.

<!-- prettier-ignore -->
  1. No trailing "-ing" phrases that add no fact ("highlighting...", "ensuring...").
  2. No vague attributions ("experts believe"). Name the source or delete.
  3. No AI vocabulary: additionally, crucial, delve, enhance, foster, landscape, pivotal, showcase, testament, underscore, vibrant.
  4. Say "is" or "has", not "serves as", "stands as", "boasts", "features".
  5. No "not just X, but Y". State the point.
  6. Don't force ideas into threes.
  7. One name per thing. Repeat it.
  8. No "from X to Y" unless X and Y sit on a real scale.
  9. No em dashes, en dashes, or hyphens used as dashes, and no parentheses in their place. Use a period or a comma.
  10. A colon goes before a list or an example, never mid-sentence as a connector.
  11. Don't bold every proper noun or acronym.
  12. No bold label and colon that restates its line ("Performance: Performance improved..."). A bold lead-in that ends in a period and adds new detail is fine.
  13. Sentence case headings.
  14. No decorative emojis.
  15. Straight quotes.
  16. No chatbot phrases ("I hope this helps!", "Let me know if...").
  17. No praise or agreement openers. Respond directly.
  18. Cut filler. "In order to" becomes "to". Delete "it is important to note that".
  19. One hedge at most: "may".
  20. No generic conclusions. State specific plans or facts.
  21. No abstract metaphor nouns: substrate, wedge, vector, nexus, primitive, harness, surface, scaffolding, paradigm, ratchet, north star, flywheel. Use the concrete word.
  22. Say what it does, not how it feels. Name the mechanism or the number. Cut a sentence that could appear unchanged in another project's docs.
  23. One idea per sentence. Split a sentence the reader has to reread.
  24. Active voice. Name the actor unless it is unknown or doesn't matter.
  25. Cut adverbs or use a stronger verb. Give the number.
  26. Use the plain word: "use", not "utilize" or "leverage".
  27. No metaphor, aphorism, personified code, or figurative verb where a literal phrase exists.
  28. Write whole sentences with their articles and verbs. Spell out arrows and abbreviations.

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/christofferbergj/agent-skills/unslop">View unslop on skillZs</a>