Strip AI writing patterns from text — em dashes, stock phrases, promotional inflation, performed authenticity, rule-of-three lists. Use when user says "humanize this", "make it sound human", "strip AI patterns", "clean up the copy", or after /content-gen or /landing-gen produces output.
Strip AI writing patterns from user-facing text. Takes a file or pasted text and rewrites it to read like a human wrote it, without losing meaning or structure.
LLM output has recognizable tells — em dashes, stock phrases, promotional inflation, performed authenticity. Readers (and Google) notice. This skill catches those patterns and rewrites them.
/content-gen, /landing-gen, /video-promo — polish the output/humanize path/to/file.md$ARGUMENTS — reads and rewrites in place.md, .txt, and text content in .tsx/.html (string literals only)The most obvious AI tell. Replace with commas, periods, colons, or restructure the sentence.
| Before | After | |--------|-------| | "The tool — which is free — works great" | "The tool (which is free) works great" | | "Three features — speed, security, simplicity" | "Three features: speed, security, simplicity" | | "We built this — and it changed everything" | "We built this. It changed everything." |
Rule: Max 1 em dash per 500 words. Zero is better.
Phrases that signal "AI wrote this." Remove or replace with specific language.
Filler phrases (delete entirely):
Promotional inflation (replace with specifics):
Performed authenticity (rewrite):
AI loves triplets: "fast, secure, and scalable." Real writing varies list length.
| Before | After | |--------|-------| | "Fast, secure, and scalable" | "Fast and secure" (if scalable isn't proven) | | "Build, deploy, and iterate" | "Build and ship" (if that's what you mean) | | Three bullet points that all say the same thing | One clear bullet |
Rule: If you find 3+ triplet lists in one document, break at least half of them.
Every section has the same shape: AI tends to write: heading → one-sentence intro → 3 bullets → transition sentence. Real writing varies section length and structure.
Hedging sandwich: "While X has limitations, it offers Y, making it Z." → Pick a side. State it.
False balance: "On one hand X, on the other hand Y." → If one side is clearly better, say so.
Word-level fixes miss these. They survive shortening: a 300-character message can still carry three of them, which is why manually trimming text does not make it sound human. Cut them, don't rewrite them into better versions.
| Pattern | Before | After |
|---------|--------|-------|
| Binary contrast | "It's not a docs problem. It's a coverage problem." | "6 questions fail because 3 devices have no docs." |
| Negative listing | "Not a wrapper. Not a chatbot. A compiler." | "It's a compiler." |
| Colon reveal | "The catch: it needs your docs." | "It needs 3 docs nobody has written." |
| Faux-insight setup | "The part everyone misses is distribution." | "Distribution is the moat." |
| Throat-clearing | "Here's the thing." / "Let me be clear." | (cut, state the point) |
| Rhetorical setup | "What if I told you...", "Think about it:" | (cut, make the claim) |
| -ing pseudo-analysis | "…shipped v2, highlighting their focus on speed." | "…shipped v2, which cut load time to 400ms." |
| Importance puffery | "marks a pivotal moment", "a testament to" | state the fact, let the reader judge |
| Metadiscourse | "This matters more than it sounds." | (cut — the fact carries it) |
| Weasel attribution | "experts agree", "studies show" | name the source or drop the claim |
| Synonym cycling | "the agent… the assistant… the tool…" | one name, repeated |
| Profound kicker | "The future isn't coming. It's already here." | delete it, end on the last concrete line |
| Summary recap | "In conclusion…", "Overall…" | end on the takeaway or next action |
Portability test. If a sentence would read identically about another person, company, or product, it is filler. Replace it with a fact, mechanism, number or consequence specific to this subject, or cut it.
Show, don't label. Cut commentary that tells the reader a point is important, surprising or subtle. If the prose already shows it, delete the label.
Protect the specific fact. Never smooth a detail into generic importance. "significantly improves productivity" → "cut review time from 30 minutes to 8."
Read the input — file path or pasted text.
Scan for patterns — check each category above. Count violations per category.
Rewrite — fix each violation while preserving:
Report what changed:
Humanized: {file or "pasted text"}
Changes:
Em dashes: {N} removed
Stock phrases: {N} replaced
Inflation: {N} deflated
Triplets: {N} broken
Sycophancy: {N} cut
Rhetorical patterns: {N} cut
Total: {N} patterns fixed
Before: {word count}
After: {word count}
If file path: write the cleaned version back. Show a diff summary. If pasted text: output the cleaned version directly.
Asked "is this AI slop?" or handed someone else's draft to audit: name each pattern from this skill that appears, quote the line, give the fix in a few words. Do not rewrite, do not score the text, and never claim an AI wrote it — detectors guess, a named pattern is evidence the writer can check. Offer to edit afterwards.
The rhetorical-pattern taxonomy in section 7 and the detect-mode contract are adapted from no-ai-slop by Peter Yang (MIT).
Google Workspace CLI for Gmail, Calendar, Drive, Contacts, Sheets, and Docs.
Manage Apple Notes via the `memo` CLI on macOS (create, view, edit, delete, search, move, and export notes). Use when a user asks OpenClaw to add a note, list notes, search notes, or manage note folders.
Work with Obsidian vaults (plain Markdown notes) and automate via obsidian-cli.
Use when you need to control Slack from OpenClaw via the slack tool, including reacting to messages or pinning/unpinning items in Slack channels or DMs.
Manage Apple Reminders via remindctl CLI (list, add, edit, complete, delete). Supports lists, date filters, and JSON/plain output.
Manage Trello boards, lists, and cards via the Trello REST API.
Category:productivity