agentleFS
cybertronaicybertronai/SutroYaro/.claude/skills/anti-slop-guide/SKILL.md

anti-slop-guide

Skill16 starsChanged 6 months agoUnlicenseView on GitHub

Use when drafting, editing, or reviewing any prose to detect and remove AI writing patterns including overused vocabulary (delve, tapestry, landscape), formulaic structures (binary contrasts, rule of three), throat-clearing openers, business jargon, and other LLM tells

The file

Show its rules and text

We split this skill into the rules it gives an agent

19 rules. About 0 are common, 19 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈19

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Cut filler phrases: Remove throat-clearing openers and emphasis crutches.

    line 14
  2. Distinctive:

    Break formulaic structures: Avoid binary contrasts, dramatic fragmentation, rhetorical setups.

    line 15
  3. Distinctive:

    Cut quotables: If it sounds like a pull-quote or motivational poster, rewrite it.

    line 18
  4. Distinctive:

    Prefer plain verbs: Use "said" not "emphasized," "used" not "leveraged," "showed" not "showcased."

    line 20
  5. Distinctive:

    Use simple copulatives: Write "X is Y" instead of "X serves as Y" or "X stands as Y."

    line 21
  6. Distinctive:

    Kill the em-dash habit: Use commas or periods where LLMs reflexively reach for em dashes.

    line 22
  7. Distinctive:

    Remove these. State the content directly.

    line 71
  8. Distinctive:

    Fix: Use two items or one. Break the triple pattern.

    line 280
  9. Distinctive:

    Only use an adjective if it adds concrete information about scale, performance, or constraints

    line 427
  10. Distinctive:

    Before delivering prose, check:

    line 435
  11. Distinctive:

    Prefer paragraphs over bullet lists for explanation

    line 547
  12. Distinctive:

    Use bullet lists only for truly distinct items (steps, comparisons)

    line 548
  13. Distinctive:

    Use subheadings sparingly, not for every paragraph

    line 549
  14. Distinctive:

    Keep headings short and factual, not dramatic two-part constructions

    line 550
  15. Distinctive:

    No analogies unless providing non-obvious clarification

    line 553
  16. Distinctive:

    Never "Imagine..." or "Think of it like this..."

    line 555
  17. Distinctive:

    No symmetric five-paragraph essay structure

    line 559
  18. Distinctive:

    Start with content, not with a question

    line 561
  19. Distinctive:

    No more than one rhetorical question per piece (prefer zero)

    line 564

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Anti-Slop Guide

Field guide for eliminating AI writing patterns from prose. Compiled from multiple sources including Wikipedia's Signs of AI Writing, community research, and practitioner guides.


Part 1: Core Rules

  1. Cut filler phrases. Remove throat-clearing openers and emphasis crutches. Say the thing directly.
  2. Break formulaic structures. Avoid binary contrasts, dramatic fragmentation, rhetorical setups.
  3. Vary rhythm. Mix sentence lengths. Two items beat three. End paragraphs differently each time.
  4. Trust readers. State facts directly. Skip softening, justification, hand-holding.
  5. Cut quotables. If it sounds like a pull-quote or motivational poster, rewrite it.
  6. Be specific. Replace generic statements with concrete details. Specificity is the antidote to slop.
  7. Prefer plain verbs. Use "said" not "emphasized," "used" not "leveraged," "showed" not "showcased."
  8. Use simple copulatives. Write "X is Y" instead of "X serves as Y" or "X stands as Y."
  9. Kill the em-dash habit. Use commas or periods where LLMs reflexively reach for em dashes.
  10. Never triple. Two items in a list, not three. Three-item lists are the most reliable AI tell.

Part 2: Overused "AI Vocabulary" Words

These words started appearing far more frequently in text after 2023. They co-occur in LLM output; where there is one, there are likely others. Introducing lots of them is one of the strongest tells for AI use.

Tier 1: Strongest Tells (almost certainly AI if clustered)

additionally, align with, crucial, delve, emphasizing, enduring, enhance, fostering, garner, highlight (as a verb), interplay, intricate/intricacies, key (as an adjective), landscape (abstract noun), pivotal, showcase, tapestry (abstract noun), testament, underscore (as a verb), valuable, vibrant

Tier 2: Strong Tells (very common in AI output)

  • bustling, comprehensive, multifaceted, nuanced, paramount
  • realm, robust, seamless/seamlessly, synergy, transformative
  • commendable, cornerstone, groundbreaking, innovative, noteworthy
  • spearhead, streamline, testament to, trailblazing, unwavering

Tier 3: Suspicious in Clusters

  • bolster, burgeoning, captivate, cater to, cognizant
  • compelling, culminate, delineate, demystify, discern
  • elucidate, embark, encompass, envisage, epitomize
  • equip, evolving, exemplify, expedite, facilitate
  • fathom, forge, foster, grapple, harness
  • illuminate, imbue, imperative, indispensable, ingenious
  • juxtapose, meticulous, mitigate, navigate, nestle
  • optimize, orchestrate, paradigm, perpetuate, pervasive
  • plethora, poignant, proliferate, propel, quintessential
  • reimagine, resonate, revolutionize, rigorous, salient
  • scrutinize, sentiment, testament, transcend, traverse
  • underscore, unprecedented, unravel, utilize, volition

AI-Overused Intensifiers

deeply, truly, fundamentally, inherently, simply, literally, inevitably, remarkably, notably, particularly, essentially, undeniably, profoundly

Absolute Words (false authority)

always, never, everyone, everybody, nobody, undoubtedly, unquestionably


Part 3: Phrases to Remove

Throat-Clearing Openers

Remove these. State the content directly.

  • "Here's the thing:"
  • "The uncomfortable truth is"
  • "It turns out"
  • "The real [X] is"
  • "Let me be clear"
  • "The truth is,"
  • "I'll say it again:"
  • "I'm going to be honest"
  • "Can we talk about"
  • "Here's what I find interesting"
  • "Here's the problem though"
  • "But here's the thing"
  • "Something shifted"
  • "Everything changed"

Emphasis Crutches

These add no meaning. Delete them.

  • "Full stop." / "Period."
  • "Let that sink in."
  • "This matters because"
  • "Make no mistake"
  • "Here's why that matters"
  • "Think about it."
  • "Read that again."
  • "I cannot stress this enough"

Generic Essay Openers

  • "In today's fast-paced world..."
  • "In today's [X]..."
  • "As we navigate the complexities..."
  • "In a world where..."
  • "As technology continues to evolve..."
  • "In the ever-evolving landscape of..."

Generic Essay Closers

  • "In conclusion..."
  • "In summary..."
  • "Overall..."
  • "As [X] continues to evolve..."
  • "The future of [X] is bright"
  • "Only time will tell"

Meta-Commentary

Remove self-referential asides:

  • "Hint:"
  • "Plot twist:" / "Spoiler:"
  • "You already know this, but"
  • "But that's another post"
  • "X is a feature, not a bug"
  • "Dressed up as"
  • "In this section, we'll look at..."
  • "As mentioned earlier..."
  • "Now that we've explored X..."
  • "First, we will examine... then we will explore..."

Performative Emphasis

False intimacy or manufactured sincerity:

  • "creeps in"
  • "I promise"
  • "They exist, I promise"
  • "This is genuinely hard"
  • "This is what leadership actually looks like"
  • "This is what X actually looks like"

Filler Adverbs and Transitions

Cut or replace:

  • "At its core"
  • "It's worth noting"
  • "Interestingly,"
  • "Importantly,"
  • "Crucially,"
  • "At the end of the day"
  • "When it comes to"
  • "The reality is"
  • "Furthermore," / "Moreover," / "In addition," (when stacked)
  • "It's important to note/remember/consider"
  • "Worth noting"

Collaborative Communication Leakage

These appear when chatbot-to-user phrasing leaks into prose:

  • "I hope this helps"
  • "Of course!"
  • "Certainly!"
  • "You're absolutely right!"
  • "Would you like..."
  • "Is there anything else"
  • "Let me know"
  • "Here is a..."
  • "Great question!"

Politeness Padding

  • "I hope this email finds you well..."
  • "I would like to take a moment to..."
  • "I was wondering if..."
  • "I wanted to kindly follow up regarding..."
  • Excessive "just," "hopefully," "perhaps"

Part 4: Business Jargon Replacements

AvoidUse instead
Navigate (challenges)Handle, address, deal with
Unpack (analysis)Explain, examine
Lean intoAccept, embrace
Landscape (context)Situation, field, area
Game-changerSignificant, important
Double downCommit, increase
Deep diveAnalysis, examination
Take a step backReconsider
Moving forwardNext, from now
Circle backReturn to, revisit
On the same pageAligned, agreed
LeverageUse
HarnessUse, apply
Unlock potentialEnable, allow
Paradigm shiftChange, shift
SynergyCooperation, combined effect
StreamlineSimplify
RobustStrong, reliable
SeamlessSmooth
Cutting-edgeNew, advanced
Best-in-classLeading, top
GroundbreakingNew, original

Part 5: Structural Patterns to Avoid

Binary Contrasts

These create false drama. State the point directly.

PatternProblem
"Not because X. Because Y."Telegraphed reversal
"[X] isn't the problem. [Y] is."Formulaic reframe
"The answer isn't X. It's Y."Predictable pivot
"It feels like X. It's actually Y."Setup/reveal cliche
"The question isn't X. It's Y."Rhetorical misdirection
"Not X. But Y."Mechanical contrast
"It's not this. It's that."Same formula, different words
"stops being X and starts being Y"False transformation arc
"It isn't just X, it's Y"Promotional reframe
"X is more than just Y; it's Z"Escalation cliche
"It wasn't X, it was Y"Hindsight reversal
"This is where X comes in"Manufactured entrance

Fix: State Y directly. "The problem is Y." "Y matters here."

Dramatic Fragmentation

Sentence fragments for emphasis read as manufactured profundity.

PatternProblem
"[Noun]. That's it. That's the [thing]."Performative simplicity
"X. And Y. And Z."Staccato drama
"This unlocks something. [Word]."Artificial revelation

Fix: Complete sentences. Trust content over presentation.

Rhetorical Setups

These announce insight rather than deliver it.

PatternProblem
"What if [reframe]?"Socratic posturing
"Here's what I mean:"Redundant preview
"Think about it:"Condescending prompt
"And that's okay."Unnecessary permission
"So, what exactly is X?"Fake curiosity
"But now? You won't believe..."Manufactured suspense

Fix: Make the point. Let readers draw conclusions.

Negative Parallelisms

Parallel constructions involving "not", "but", or "however" such as "Not only ... but ..." or "It is not just about ..., it's ..." are very common in LLM writing to appear balanced and thoughtful.

  • "not ..., it's ..."
  • "no ..., no ..., just ..."
  • "Not only X, but also Y"

Rule of Three

LLMs overuse the rule of three. This takes different forms:

  • "adjective, adjective, adjective"
  • "short phrase, short phrase, and short phrase"
  • Three-item lists to make superficial analyses appear comprehensive
  • Three examples where one or two would suffice

Fix: Use two items or one. Break the triple pattern.

False Ranges

LLMs employ figurative "from ... to ..." constructions where the endpoints are loosely related or unrelated and no meaningful scale can be inferred.

  • "from boardrooms to battlefields"
  • "from grassroots to global"
  • "from problem-solving to scientific discovery"

Fix: Be specific about what you actually mean.

Formulaic Constructions

PatternProblem
"By the time X, I was Y."Narrative template
"X that isn't Y"Indirect. Say "X is broken"
"Despite its [positive], X faces challenges..."Outline-like conclusion
"Challenges and Legacy"AI section heading cliche
"Future Outlook" / "Future Prospects"Speculative padding

The "Challenges" Formula

Many AI articles include a section that begins "Despite its [positive quality], [subject] faces challenges..." and ends with vaguely positive speculation. This rigid formula is a strong AI tell.


Part 6: Rhythm and Style Patterns

Sentence-Level Issues

PatternFix
Three consecutive sentences match lengthBreak one
Paragraph ends with punchy one-linerVary endings
Em-dash before a revealRemove it, use period or comma
Explaining a metaphorTrust it to land
Questions answered immediatelyLet questions breathe or cut them
Paragraphs starting with "So"Start with content
Sentences starting with "Look,"Remove
Every paragraph ends punchilyVary endings
Staccato fragmentationDon't stack short punchy sentences
Dashes for dramatic pauseUse commas or periods
"Not always. Not perfectly."Hedging disguised as reassurance
Multiple sentences opening identicallyVary first 3-5 words
Uniform paragraph lengthMix short and long
Reusing "Furthermore/Moreover/In addition"Use one, cut the rest, or restructure

Formatting Tells

  • Overuse of boldface: Mechanical emphasis in a "key takeaways" fashion
  • Emoji in headings/bullets: Decorating section headings or bullet points with emoji
  • Inline-header vertical lists: Bold header followed by colon, then description text
  • Excessive bullet points: Replacing developed paragraphs with bulleted lists
  • Unnecessary tables: Small tables that could be stated as prose
  • Title Case in headings: Capitalizing All Main Words in headings (prefer sentence case)
  • Decorative symbols: Unnecessary arrows (->), check marks, or Unicode styling
  • Skipping heading levels: Jumping from H1 to H3

Copulative Avoidance

LLMs substitute constructions like "serves as a" or "marks the" for simpler "is" or "are" phrasing. One study documented an over 10% decrease in "is" and "are" usage in AI-assisted academic writing.

AI prefersHuman writes
serves asis
stands asis
marksis
representsis
boastshas
featureshas
offershas

Elegant Variation (Unnecessary Synonyms)

AI has a repetition-penalty that makes it avoid reusing words. It will give a character's name then switch to "the protagonist," "the key player," "the eponymous character" etc. If you mean the same thing, use the same word.


Part 7: Content-Level AI Tells

Undue Emphasis on Significance

LLM writing puffs up importance by adding statements about how aspects "represent" or "contribute to" broader topics.

Words to watch: stands/serves as, is a testament/reminder, a vital/significant/crucial/pivotal/key role/moment, underscores/highlights its importance/significance, reflects broader, symbolizing its, ongoing/enduring/lasting, contributing to the, setting the stage for, marking/shaping the, represents/marks a shift, key turning point, evolving landscape, focal point, indelible mark, deeply rooted

Promotional and Advertisement-Like Language

LLMs have serious problems keeping a neutral tone, especially when writing about anything that could be considered "cultural heritage."

Words to watch: boasts a, vibrant, rich (figurative), profound, enhancing its, showcasing, exemplifies, commitment to, natural beauty, nestled, in the heart of, groundbreaking (figurative), renowned

Superficial Analyses

AI inserts superficial analysis, often by attaching a present participle ("-ing") phrase at the end of sentences.

Words to watch: highlighting/underscoring/emphasizing..., ensuring..., reflecting/symbolizing..., contributing to..., cultivating/fostering... (figurative), encompassing..., valuable insights, align/resonate with

Vague Attributions and Overgeneralization

AI attributes opinions or claims to vague authority.

Words to watch: Industry reports, Observers have cited, Experts argue, Some critics argue, several sources/publications (when only few are cited), such as (before exhaustive word lists)

Knowledge-Cutoff Disclaimers

Phrases that leak the chatbot's awareness of its own limitations into prose:

  • "as of [date]"
  • "Up to my last training update"
  • "as of my last knowledge update"
  • "While specific details are limited/scarce..."
  • "not widely available/documented/disclosed"
  • "...in the provided/available sources/search results..."
  • "based on available information..."

Over-Signposting

Announcing what you're about to say instead of saying it:

  • "In this section, we will discuss..."
  • "This will include a discussion of..."
  • "The purpose is to provide a comprehensive understanding..."
  • "Now that we've explored X, let's move on to Y"

Safe Conclusions

Zooming out to high-level generalizations:

  • "As AI continues to evolve..."
  • "The future remains bright for..."
  • "Only time will tell..."
  • Recap-only endings with no specific insight
  • Paragraph that could work in any article on any topic

Part 8: Adjective Audit Process

  1. Strip all adjectives from a paragraph
  2. Restore only those that change meaning usefully
  3. Replace mood with measurement:
    • "robust system" -> "handles 10k requests/second without data loss"
    • "comprehensive solution" -> state what it actually covers
    • "innovative approach" -> describe what's new about it
  4. Only use an adjective if it adds concrete information about scale, performance, or constraints

Part 9: Two-Pass Revision Workflow

Pass 1: Self-Critique

Before delivering prose, check:

  • Three consecutive sentences match length? Break one.
  • Paragraph ends with punchy one-liner? Vary it.
  • Em-dash before a reveal? Remove it.
  • Explaining a metaphor? Trust it to land.
  • Any word from Tier 1 vocabulary list? Replace it.
  • Any binary contrast structure? Flatten it.
  • Rule of three? Cut to two or one.
  • Generic opener or closer? Cut it.
  • Same first 3-5 words on multiple sentences? Vary them.
  • Any "serves as" / "stands as" constructions? Use "is."
  • Present participle ("-ing") phrase tacked onto sentence end? Evaluate if needed.
  • Self-referential meta-commentary? Remove it.

Pass 2: Scoring

Rate 1-10 on each dimension:

DimensionQuestion
DirectnessStatements or announcements?
RhythmVaried or metronomic?
TrustRespects reader intelligence?
AuthenticitySounds human?
DensityAnything cuttable?
SpecificityConcrete or generic?

Below 40/60: revise.


Part 10: Before/After Examples

Example 1: Throat-Clearing + Binary Contrast

Before:

"Here's the thing: building products is hard. Not because the technology is complex. Because people are complex. Let that sink in."

After:

"Building products is hard. Technology is manageable. People aren't."

Example 2: Filler + Unnecessary Reassurance

Before:

"It turns out that most teams struggle with alignment. The uncomfortable truth is that nobody wants to admit they're confused. And that's okay."

After:

"Teams struggle with alignment. Nobody admits confusion."

Example 3: Business Jargon Stack

Before:

"In today's fast-paced landscape, we need to lean into discomfort and navigate uncertainty with clarity. This matters because your competition isn't waiting."

After:

"Move faster. Your competition is."

Example 4: Dramatic Fragmentation

Before:

"Speed. Quality. Cost. You can only pick two. That's it. That's the tradeoff."

After:

"Speed, quality, cost -- pick two."

Example 5: Rhetorical Setup

Before:

"What if I told you that the best teams don't optimize for productivity? Here's what I mean: they optimize for learning. Think about it."

After:

"The best teams optimize for learning, not productivity."

Example 6: Promotional Puffery (Wikipedia-style)

Before:

"Nestled within the breathtaking region of Gonder in Ethiopia, Alamata Raya Kobo stands as a vibrant town with a rich cultural heritage and a significant place within the Amhara region. From its scenic landscapes to its historical landmarks, Alamata Raya Kobo offers visitors a fascinating glimpse into the diverse tapestry of Ethiopia."

After:

"Alamata Raya Kobo is a town in the Amhara region of Ethiopia, in the Gonder area."

Example 7: Superficial Analysis with -ing Phrases

Before:

"The founding of Idescat represented a significant shift toward regional statistical independence, enabling Catalonia to develop a statistical system tailored to its unique socio-economic context. This initiative was part of a broader movement across Spain to decentralize administrative functions and enhance regional governance."

After:

"Idescat gave Catalonia its own statistical system. Spain was decentralizing administrative functions at the time."

Example 8: Overuse of Significance Words

Before:

"This etymology highlights the enduring legacy of the community's resistance and the transformative power of unity in shaping its identity."

After:

"The name reflects the community's history of resistance."


Part 11: Complete Blacklisted Word List

Words to Avoid (unless the specific technical meaning is required)

alchemy, beacon, bustling, captivate, cherished, commendable, comprehensive, cornerstone, crucible, cutting-edge, dazzling, delve, demystify, discerning, dynamic, elucidate, embark, empower, enchanting, endeavor, enduring, enigma, epitome, ever-evolving, exquisite, facilitate, fascinating, first-hand, flourish, forge, foster, game-changer, garnish, gateway, groundbreaking, hallmark, harness, holistic, ignite, illuminate, imbue, immerse, impactful, imperative, impressive, in-depth, indispensable, ingenious, innovative, insightful, inspire, instrumental, interplay, intricate, invaluable, journey (figurative), juxtapose, keen, key (adjective), landscape (figurative), leverage, luminary, masterful, meticulous, mind-blowing, monumental, multifaceted, must-read, navigate, nestle, noteworthy, nuanced, orchestrate, outstanding, paradigm, paramount, passion-driven, perpetuate, pervasive, pivotal, plethora, poignant, premier, prestigious, profound, proliferate, propel, quintessential, realm, reimagine, remarkable, renowned, resonate, revolutionize, rigorous, robust, salient, scrutinize, seamless, sentiment, showcase, significant (as filler), spearhead, staggering, streamline, superb, symbiotic, synergy, tapestry (figurative), testament, thought-provoking, thriving, trailblazing, trajectory, transformative, traverse, treasure trove, underscore, undeniable, underpin, unearth, unparalleled, unprecedented, unveil, unwavering, utilize, vibrant, visionary, vital, volition, world-class


Part 12: Structural Rules for Specific Content Types

For Technical Writing / Documentation

  • Prefer paragraphs over bullet lists for explanation
  • Use bullet lists only for truly distinct items (steps, comparisons)
  • Use subheadings sparingly, not for every paragraph
  • Keep headings short and factual, not dramatic two-part constructions
  • One consistent point of view ("we" or "you"), stick to it
  • Full sentences, no stylistic fragments
  • No analogies unless providing non-obvious clarification
  • Maximum one analogy per piece
  • Never "Imagine..." or "Think of it like this..."

For Blog Posts / Articles

  • No symmetric five-paragraph essay structure
  • Vary section lengths
  • Start with content, not with a question
  • End with specific insight, not a generic zoom-out
  • Use contractions naturally
  • No more than one rhetorical question per piece (prefer zero)

For Emails / Professional Communication

  • "Any update on...?" not "I wanted to kindly follow up regarding..."
  • Direct requests, minimal padding
  • No "I hope this email finds you well"
  • No "I would like to take a moment to"

Part 13: The Core Insight

The fundamental difference between human and AI writing:

AI deploys stylistic devices reflexively and indiscriminately. Humans use them intentionally and sparingly.

Good writing varies structure. AI maintains unnatural consistency. Human authenticity comes from specificity rooted in tangible experience and developed voice.

The AI signature is "surface polish with nothing underneath" -- paragraphs that say in four sentences what needs one, coherent individually but lacking a throughline, plausible but never specific.

The antidote is always the same: be concrete, be specific, say less.


Sources

License

MIT. Original stop-slop skill by Hardik Pandya. Compiled and expanded by Yad Konrad.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.