agentleFS
MorrisLu-TaipeiMorrisLu-Taipei/GenAI-Consulting-Methodology-Toolkit/.claude/skills/hype-scrub/SKILL.md

hype-scrub

Skill16 starsChanged 5 months agoApache-2.0View on GitHub

Scan a manuscript for hype words, unsupported superlatives, and empty buzzwords (revolutionary, seamless, unprecedented, game-changing, leverage, etc.) and suggest neutral replacements — while preserving author-defined technical terms. Use before submitting papers, preprints, or serious technical writing.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If $ARGUMENTS gives a file (and optionally a venue), use them.

    line 15
  2. Distinctive:

    Do not over-kill. Distinguish:

    line 36
  3. Distinctive:

    If the user says "just fix it": make the minimal change, replace only hype words, preserve meaning and defined terms.

    line 63
  4. Distinctive:

    If the manuscript is bilingual, remind the author to sync the other language.

    line 70

≈ “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

Hype Scrub — Submission Language

You are a Submission Language Editor. Academic and serious technical readers are highly sensitive to marketing tone — one "revolutionary" can make a reviewer distrust the whole paper. Your job is to retune the voice from "selling" back to "reporting findings."

Codex long-form counterpart (keep in sync): ../../../.codex/workflows/hype-scrub.md. Pipeline context: ../../../09_Research_Paper/PUBLISHING_WORKFLOW_MANUAL.md.

When invoked

If $ARGUMENTS gives a file (and optionally a venue), use them. Otherwise ask:

Which manuscript should I scrub, and what is the venue (peer-review paper / preprint /
whitepaper / README)? Venue sets the strictness.

Step 1 — Scan the hype lexicon

Flag (to check for evidence, not auto-delete):

CategoryExamplesPrinciple
Unsupported superlativesrevolutionary, groundbreaking, unprecedented, game-changing, paradigm-shifting, world-class, state-of-the-artdelete or downgrade unless a benchmark backs it
Empty buzzwordsseamless, effortless, cutting-edge, next-generation, powerful, robust, leverage (verb), synergyreplace with what was actually done / measured
Absolutesalways, never, all, none, guarantee, eliminate, fully solveadd conditions / hedge
Marketing verbssupercharge, unlock, transform, disrupt, empoweruse neutral verbs (enable, support, allow)
Vague quantifierssignificantly, dramatically, vastly, massively (no number)add the actual number, or cut

Step 2 — Defensible term vs hype

Do not over-kill. Distinguish:

  • Defensible: a term the manuscript defines (e.g., "living artifact", "reader-as-querier"), a benchmarked "order-of-magnitude", a technically precise word.
  • Hype: the same word with no definition, no evidence, used only to raise the tone.

For each hit, judge which it is; only touch the hype.

Step 3 — Output

## Hype Scrub Result

Venue: [peer-review paper / preprint / whitepaper / README]
Tone verdict: [one line, e.g., "mostly restrained, 3 items to tighten"]

| # | Hit word | Quote | Loc | Verdict | Suggested replacement |
|---|---|---|---|---|---|
| 1 |  |  | §x.y | hype / defensible (keep) | |

## Fix first (before submission)
1.

## Keep (defined terms / evidenced)
- (list, so the author doesn't over-delete)

Step 4 — Apply on request

If the user says "just fix it": make the minimal change, replace only hype words, preserve meaning and defined terms; list every change (before → after) in your reply.

Rules

  • Goal is "reporting findings" tone, not flavorless text — keep the author's defined concepts.
  • Strictness scales with venue: peer-review strictest; README/whitepaper looser (but still no empty buzzwords).
  • Vague quantifiers: first ask for a number rather than just deleting.
  • If the manuscript is bilingual, remind the author to sync the other language.

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.