Humanizer

What You Get A complete SKILL.md that teaches your OpenClaw agent to detect and remove 24 documented AI writing patterns, including: • AI Vocabulary — "delve," "pivotal," "tapestry," "landscape," "underscore," and 20+ more high-frequency tells • Inflated Significance — "marks a pivotal moment," "testament to," "underscores its importance" • Vague Attributions — "experts argue," "industry reports suggest," "observers have noted" • Structural Tells — rule of three, negative parallelisms ("it's not just X, it's Y"), em dash overuse • Promotional Puffery — "nestled," "vibrant," "breathtaking," "must-visit," "boasts" • Sycophantic Openers — "Great question!" "I'd be happy to help!" "Absolutely!" • Soulless Writing — technically clean but flat, voiceless, no opinions, no rhythm

How It Works

1. Pass any text to your agent with a humanize instruction
2. The agent scans for all 24 pattern categories
3. Rewrites problem sections while preserving meaning and matching your intended tone
4. Adds voice — opinions, varied rhythm, first person where it fits, specific details over vague claims

Why This Exists AI writing is everywhere now and readers can feel it instantly — even when they can't name what's wrong. Stripping the patterns is only half the job. The other half is making sure something with a pulse replaces them. This skill does both.

Built on Wikipedia's "Signs of AI Writing" guide maintained by WikiProject AI Cleanup — the most comprehensive public catalog of AI writing patterns available.

Customization Works on any content type: blog posts, emails, proposals, social copy, support responses, reports. Pass a tone instruction ("keep it formal" or "conversational") and the agent matches it. The pattern list is documented in the skill file so your agent can explain every edit it makes.