Skip to content
Back to search

skill-deslop

De-AI-ify scientific writing

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/stephenturner--skill-deslop/badge.svg)](/listings/stephenturner--skill-deslop)

HTML

<a href="/listings/stephenturner--skill-deslop"><img src="/api/listings/stephenturner--skill-deslop/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

**Option 1: Download ZIP**

Click the green **Code** button at the top of this repo, then **Download ZIP**. Extract the ZIP and add the folder to your Claude skills directory.

**Option 2: Releases**

Go to the [Releases](https://github.com/stephenturner/skill-deslop/releases) page and download the latest `.skill` file. Add it to your Claude skills in [customize/skills](https://claude.ai/customize/skills) on the web, or double-click it if you have Claude Desktop installed.

**Option 3: Build it yourself**

Build a `.skill` file from the source code, then add it to your Claude skills as described above.

git clone https://github.com/stephenturner/skill-deslop.git
cd skill-deslop
zip -r deslop.skill SKILL.md references/

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

SKILL.md · Manifest · 130 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ---
  2. name: deslop
  3. description: Remove AI writing patterns from prose. Use this skill when writing, drafting, editing, reviewing, or revising any text to eliminate predictable AI tells, slop, and formulaic patterns. Trigger this skill whenever the user asks to "deslop", "de-AI", "make it sound human," "remove AI patterns," "remove AI tropes," "clean up AI writing," fix "slop," "deslop" text, or review prose for authenticity. Also use when the user asks you to write or draft anything and wants it to sound natural rather than AI-generated. Common use cases include scientific writing (manuscripts, abstracts, cover letters, grant narratives, discussion sections, peer review responses), blog posts, newsletters, memos, reports, and any other substantial prose.
  4. ---
  5. # Deslop: Remove AI Writing Patterns from Prose
  6. Strip predictable AI patterns from writing. Make prose sound like a specific human wrote it, not like a language model generated it.
  7. ## When to Apply
  8. - Any request to "make it sound human" or "deslop" writing
  9. - Any prose (articles, blog posts, essays, memos, newsletters, reports) or scientific writing (manuscripts, abstracts, cover letters, grant narratives, discussion sections, peer review responses) where the user wants it to sound natural rather than AI-generated
  10. - Editing or revising existing text where the user wants it to sound natural rather than AI-generated
  11. - Reviewing text for AI tells
  12. ## Core Rules
  13. ### 1. Cut filler phrases
  14. Remove throat-clearing openers ("Here's the thing:"), emphasis crutches ("Let that sink in."), business jargon ("navigate the landscape"), and meta-commentary ("In this section, we'll explore..."). See [references/phrases.md](references/phrases.md) for the full catalog.
  15. ### 2. Break formulaic structures
  16. Avoid binary contrasts ("Not X. Y."), negative listings ("Not a X. Not a Y. A Z."), dramatic fragmentation ("Speed. That's it. That's the tradeoff."), self-posed rhetorical questions ("The result? Devastating."), and anaphora/tricolon abuse. See [references/structures.md](references/structures.md) for patterns and fixes.
  17. ### 3. Eliminate AI tropes
  18. Watch for the full catalog of AI writing tells: "quietly" and other magic adverbs, "delve" and its cousins, the "serves as" dodge, false ranges ("from X to Y" where the range is meaningless), superficial participle analyses ("highlighting its importance"), invented concept labels ("the supervision paradox"), grandiose stakes inflation, patronizing analogies, and false vulnerability. See [references/tropes.md](references/tropes.md) for the complete list with examples.
  19. ### 4. Use active voice with human subjects
  20. Prefer active constructions with named actors. "The complaint becomes a fix" is wrong. "The team fixed it" is right. If no specific person fits, use "we" in scientific prose or "you" in blog posts.
  21. ### 5. Be specific
  22. No vague declaratives ("The reasons are structural"). Name the specific thing. No lazy extremes ("every," "always," "never") doing vague work. No vague attributions ("Experts argue..."). If you cannot name the expert, you do not have a source.
  23. In scientific writing, domain terminology is fine and expected. "Weighted interval score" is precise language, not jargon. The problem is business buzzwords ("leverage," "landscape," "ecosystem") and AI vocabulary tells ("delve," "tapestry," "nuanced") leaking into technical prose.
  24. ### 6. Match register to context
  25. In blog posts and newsletters, put the reader in the room. "You" beats "People." Specifics beat abstractions. No narrator-from-a-distance voice.
  26. In scientific writing, maintain appropriate formality. Use "we" for your own work, cite specific authors instead of "researchers have shown," and avoid both the distant narrator ("It has long been recognized that...") and the overly casual blog voice. State claims and back them with citations.
  27. ### 7. Vary rhythm
  28. Mix sentence lengths. Two items beat three. End paragraphs differently. No em dashes. Do not stack short punchy fragments for manufactured emphasis. Do not write listicles disguised as prose ("The first wall... The second wall...").
  29. ### 8. Trust readers
  30. State facts directly. Skip softening, justification, hand-holding. No "Let's break this down." No "Think of it as..." No pedagogical voice unless the audience genuinely needs it. No fractal summaries (telling the reader what you are about to say, saying it, then summarizing what you said).
  31. ### 9. Watch formatting tells
  32. No bold-first bullets (every list item starting with a bolded keyword). No unicode arrows. No em dashes. No signposted conclusions ("In conclusion..."). No "Despite these challenges..." formulas. These are strong AI signals.
  33. ### 10. Do not dilute
  34. One point per section. Do not restate the same argument in ten different ways across thousands of words. Do not beat a single metaphor to death. Do not stack historical analogies for false authority ("Apple didn't build Uber. Facebook didn't build Spotify...").
  35. ## Quick Checks
  36. Run these before delivering any prose:
  37. - Heavy use of adverbs or -ly words? Cut them.
  38. - Any passive voice? Find the actor, make them the subject.
  39. - Inanimate thing doing a human verb? Name the person.
  40. - Any "here's what/this/that" throat-clearing? Cut to the point.
  41. - Any "not X, it's Y" contrasts? State Y directly.
  42. - Any self-posed rhetorical question answered immediately? Fold into a statement.
  43. - Three consecutive sentences match length? Break one.
  44. - Paragraph ends with a punchy one-liner? Vary it.
  45. - Em dash anywhere? Remove it. Use a comma or period or a parenthetical.
  46. - Vague declarative ("The implications are significant")? Name the specific implication.
  47. - Any sentence starting with What/When/Where/Which/Who/Why/How as a crutch? Restructure.
  48. - Meta-joiners ("The rest of this essay...")? Delete.
  49. - "It's worth noting" or similar filler transitions? Delete.
  50. - Same metaphor used more than twice? Replace or cut repeats.
  51. - "Despite these challenges..." formula? Rewrite.
  52. - Bold-first bullet pattern? Remove bold leads.
  53. - Tricolon (three-item list)? Use two items or one.
  54. ## Scoring
  55. When reviewing text, rate 1-10 on each dimension:
  56. | Dimension | Question |
  57. |-----------|----------|
  58. | Directness | Statements or announcements? |
  59. | Rhythm | Varied or metronomic? |
  60. | Trust | Respects reader intelligence? |
  61. | Authenticity | Sounds like a specific human wrote it? |
  62. | Density | Anything cuttable? |
  63. Below 35/50: revise.
  64. ## Reference Files
  65. Consult these for detailed catalogs when writing or editing:
  66. - [references/phrases.md](references/phrases.md): Phrases to remove or replace (throat-clearing, emphasis crutches, business jargon, adverbs, meta-commentary, vague declaratives)
  67. - [references/structures.md](references/structures.md): Structural patterns to avoid (binary contrasts, negative listings, dramatic fragmentation, rhetorical setups, false agency, passive voice, rhythm problems)
  68. - [references/tropes.md](references/tropes.md): Full catalog of AI writing tropes (word choice, sentence structure, paragraph structure, tone, formatting, composition)
  69. - [references/examples.md](references/examples.md): Before/after transformations showing how to fix common patterns
  70. ## Examples
  71. See [references/examples.md](references/examples.md) for before/after transformations.
  72. **Quick inline example (scientific writing):**
  73. Before:
  74. > "It's worth noting that these findings have important implications for how we navigate the challenges of forecast ensembling moving forward. Despite these challenges, this work contributes meaningfully to the growing body of literature, highlighting the need for continued evaluation."
  75. After:
  76. > "If individual model rankings are unstable across geography and time, ensemble methods that weight models by past performance may not improve on equal-weight approaches."
  77. Changes: Replaced filler transition, vague declarative, "despite these challenges" formula, and superficial participle analysis with the specific implication.
  78. **Quick inline example (blog post):**
  79. Before:
  80. > "Here's the thing: most bioinformatics pipelines break in production. Not because the code is bad. Because the data is bad. Let that sink in."
  81. After:
  82. > "Most bioinformatics pipelines break in production. The code runs fine. The data doesn't match the assumptions baked into it."
  83. Changes: Removed opener, binary contrast, and emphasis crutch. Named the specific problem.
README.md · README · 81 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # deslop
  2. A Claude skill for removing AI writing patterns from prose.
  3. ## What it does
  4. When you write, draft, edit, or review text, `deslop` identifies and eliminates predictable AI tells: formulaic sentence structures, filler phrases, false agency, dramatic fragmentation, vague declaratives, and dozens of other patterns that signal machine-generated writing.
  5. The skill works across any prose context. Examples are weighted toward scientific writing and technical blog posts, but the rules apply to any writing where you want a human voice without the AI veneer. For scientific contexts specifically, the skill accounts for conventions like passive voice in methods sections and domain-specific terminology.
  6. ## Installation
  7. **Option 1: Download ZIP**
  8. Click the green **Code** button at the top of this repo, then **Download ZIP**. Extract the ZIP and add the folder to your Claude skills directory.
  9. **Option 2: Releases**
  10. Go to the [Releases](https://github.com/stephenturner/skill-deslop/releases) page and download the latest `.skill` file. Add it to your Claude skills in [customize/skills](https://claude.ai/customize/skills) on the web, or double-click it if you have Claude Desktop installed.
  11. **Option 3: Build it yourself**
  12. Build a `.skill` file from the source code, then add it to your Claude skills as described above.
  13. ```sh
  14. git clone https://github.com/stephenturner/skill-deslop.git
  15. cd skill-deslop
  16. zip -r deslop.skill SKILL.md references/
  17. ```
  18. ## How to use it
  19. Install as described above, then use it like you normally talk to Claude. The skill triggers automatically when you:
  20. - Ask Claude to write prose (blog posts, essays, articles, memos, newsletters) making it sound natural instead of AI-generated
  21. - Ask Claude to "deslop", "de-AI" or "make it sound human"
  22. - Ask Claude to check for "slop" or AI patterns
  23. You can also reference the skill directly:
  24. - "Review this draft using the deslop checklist"
  25. - "Score this text on the deslop rubric"
  26. - "Rewrite this paragraph to pass the deslop quick checks"
  27. ## Scoring rubric
  28. The skill includes a 1-10 scoring rubric across five dimensions:
  29. | Dimension | Question |
  30. |-----------|----------|
  31. | Directness | Statements or announcements? |
  32. | Rhythm | Varied or metronomic? |
  33. | Trust | Respects reader intelligence? |
  34. | Authenticity | Sounds like a specific human wrote it? |
  35. | Density | Anything cuttable? |
  36. Below 35/50: revise.
  37. ## Skill structure
  38. ```
  39. deslop/
  40. ├── SKILL.md # Core rules, quick checks, scoring rubric
  41. ├── README.md # This file
  42. └── references/
  43. ├── phrases.md # Phrases to remove or replace
  44. ├── structures.md # Structural patterns to avoid
  45. ├── tropes.md # Full catalog of AI writing tropes
  46. └── examples.md # Before/after transformations
  47. ```
  48. ## Acknowledgments
  49. This skill was built in part by combining and synthesizing material from two open sources:
  50. - **AI writing tropes catalog** from [tropes.fyi](https://tropes.fyi/) by [Ossama Hassanein](https://ossama.is). The `references/tropes.md` file is adapted from this source, and trope patterns are integrated throughout the other reference files.
  51. - **stop-slop** from [github.com/hardikpandya/stop-slop](https://github.com/hardikpandya/stop-slop) by [Hardik Pandya](https://hvpandya.com). The phrase lists, structural patterns, before/after examples, scoring rubric, and quick checks draw from this project.
  52. ## License
  53. MIT

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.