All skills
elithrar avatar

/anti-slop

@a2a5f65
by Matt Silverlockelithrar/dotfiles201 stars
15

Edit formulaic or AI-sounding prose while preserving the author's voice. Use for an anti-slop pass, editorial review, or requests to make writing sound more human; ordinary drafting does not require this skill.

Use this Skill: https://skilld.dev/gh/elithrar/dotfiles/anti-slop

This session only. Nothing lands on disk.

referencestells.md

≈1.9k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Editing patterns and examples

Use for a detailed editorial pass or a disputed phrasing change. These patterns are candidates, not proof of AI authorship or a word blacklist.

The tell catalog

Tell Why it erodes trust Fix
Meta throat-clearing — announcing the point before making it: "It is worth noting that…", "There is a failure mode worth naming…", "Here is the tension." Wastes the reader's time; signals padding. Delete the frame. State the thing.
Demonstrative kicker — a vague "This/That + verdict" fragment tacked after a sentence: "That instinct backfires.", "This is where the risk hides.", "That is the whole point." Formulaic filler rhythm; the "verdict" usually restates the prior sentence. Cut it, or fold the idea into the sentence; pivot the next sentence with a real transition ("But…").
Puffery adverbs — genuinely, truly, actually, simply, deeply, structurally, fundamentally. Intensifiers that add heat, not light. Delete. If the claim needs the adverb to be true, the claim is weak.
Importance-flagging — "Speed is not a footnote here.", "This matters.", "Make no mistake." Tells the reader what to feel instead of earning it. Show the consequence directly.
Clever metaphor flourish — "an authorization bug wearing the costume of a performance optimization." Reads as a model performing wit. State it plainly: "an authorization bug, not a performance optimization."
Grandiose prediction — "will define the next decade", "changes everything", "the future of X". Overclaiming; lowers credibility in serious writing. Cut or scope to a concrete, defensible claim.
Rule-of-three reflex — every list padded to three parallel items. Predictable cadence; the third item is often filler. Vary list length. Keep two if two is true.
Antithesis overuse — "not X, but Y" as a tic in paragraph after paragraph. One is rhetoric; five is a template. Keep the load-bearing one; rewrite the rest as plain statements.
Correlative bloat — "not only… but also", "whether… or". Scaffolding that inflates simple sentences. Split or simplify.
Hedged confidence — "It's important to consider…", "One might argue…", "In many ways…". Sounds authoritative while committing to nothing. Take the position or cut the sentence.
Section-closing summary — a final sentence that restates what the paragraph just said. Redundant; a model habit. Delete if it adds nothing.
Overused lexicon — delve, tapestry, realm, landscape, underscore, leverage, seamless, robust, crucial, pivotal, testament, navigate (figurative), foster, elevate, unlock. Often overused vocabulary; assess it in context. Swap for plain words or cut. See the lexicon below.

Concrete before / after

Meta throat-clearing

  • Before: "There is a failure mode worth naming directly, because well-intentioned teams walk straight into it. It is the belief that safety means a human approving every step."
  • After: "Many teams equate safety with a human approving every step."

Demonstrative kicker

  • Before: "…reuse the machinery we already have. That instinct is where most of the risk hides. Four properties break the assumptions."
  • After: "…reuse the machinery we already have. But four properties break the assumptions."

Importance-flagging + staccato

  • Before: "Speed is not a footnote here. It changes which controls are viable. Anything that depends on a person noticing has already lost."
  • After: "Speed changes which controls are viable: anything that depends on a person noticing has already lost."

Clever flourish

  • Before: "an authorization bug wearing the costume of a performance optimization"
  • After: "an authorization bug, not a performance optimization"

What to keep (do not "fix" these)

These are voice, not slop. Deleting them is the over-correction failure.

  • Earned fragments: "Not flagged after the fact. Removed." — a short beat after a long sentence, used sparingly, for emphasis.
  • Deliberate parallelism: "You can have either. You cannot have both."
  • A strong closing line: "The perimeter fell twelve years ago. The session is next."
  • First-person conviction: "We think…", "We are not comfortable saying…" when the author owns a real position.

The test: could the author defend this choice if asked? If yes, keep it. Slop cannot be defended — it is there because a pattern put it there.

Overused AI lexicon

Swap for plain words or cut. Not banned outright — flagged for scrutiny. If the plain word is worse, keep the original.

Flagged word Usually means Plainer choice
delve (into) look at, examine look at, dig into
tapestry / landscape / realm (filler noun) cut, or name the actual thing
underscore / highlight show, stress show
leverage (verb) use use
seamless / frictionless easy, smooth easy, or say how
robust strong, reliable reliable, sturdy
crucial / pivotal / vital important important, or say why
testament to shows shows
navigate (figurative) handle, deal with handle
foster / cultivate build, grow build
elevate / unlock / empower improve, enable improve, let
myriad / plethora many many, or a number
in today's (fast-paced) world (throat-clearing opener) cut
when it comes to for, about for
it's worth noting / it should be noted (padding) cut
at the end of the day (filler) cut
that being said but, still but
a double-edged sword (cliché) name the specific tradeoff
game-changer / revolutionary (overclaim) say the concrete change

Structural tells beyond word choice

  • Uniform paragraph length: every paragraph 3–4 sentences. Vary it; some ideas need one sentence.
  • Uniform sentence length: no short sentences among the long ones, or vice versa. Rhythm needs contrast.
  • Every section opens with a definition: "X is a…". Sometimes, but not every time.
  • Bullet lists where prose is better: models default to lists. Use prose for argument, lists for genuinely parallel items.
  • Symmetrical hedging: "On one hand… on the other hand…" when the author actually has a view.
  • Excessive signposting: "First… Second… Finally…" on short passages that do not need it.
  • Bold-everywhere emphasis: bolding so many phrases that none stands out.
  • Em-dash overuse: em-dashes as the default connector in sentence after sentence. Vary with commas, colons, periods, parentheses. (Some authors dislike em-dashes entirely — check preference.)
  • Tidy false balance: presenting a "both sides" wrap-up on a question the piece already answered.

The validation questions

For any flagged phrase, ask in order: 1. Does cutting it lose meaning? If yes, it may be content, not slop. Keep or reword to preserve the meaning. 2. Is it formula or choice? Would the author defend it, or did a pattern put it there? 3. Does it announce rather than deliver? Meta-commentary about the writing is almost always cuttable. 4. Does it add heat or light? Intensifiers and flourishes add heat. Cut heat, keep light. 5. Would a skeptical reader trust the sentence more or less without it? Optimize for trust.

If a phrase survives all five, leave it.

Calibration

  • Light touch for a personal blog or an author with strong style: fix only clear formula and lexicon; protect voice aggressively.
  • Heavier touch for corporate/marketing copy, which tends to be slop-dense: still surgical, but more candidates will be real.
  • Match register: a playful piece keeps playful; a white paper stays measured. Do not neutralize tone toward a generic middle.

Source: SKILL.md on GitHub

No alerts26d3 checks · Risk SAFE
  • Gen Agent Trust Hub26d

    The skill is a writing assistant designed to detect and remove common AI-generated phrasing (slop) from prose. It provides editorial guidelines and a catalog of overused terms to help make writing sound more human while preserving the original author's voice. No security risks were identified.

  • Socket26d

    No alerts

  • Snyk26d

    Risk: LOW · No issues

Signed by skilld at a2a5f65. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 weeks ago.

Activeupdated 4 weeks ago

README badge

README badge for elithrar/dotfiles/anti-slop