agentleFS
CyranoBCyranoB/department-of-prose/skills/slop-check/SKILL.md

slop-check

Skill2 starsChanged 57 days agoMITView on GitHub

Review writing patterns without rewriting text. Runs an optional algorithmic SLOP scorer (0-100, based on EQBench methodology) and reports contextual findings against a versioned 36-pattern catalogue. Review only — no rewrite or humanization.

5 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions5 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 5

    1. polarsource/polar★ 10ksame name
    2. Aidan945/slop-check★ 1same name
    3. systemonster/slop-check★ 1same name
    4. tslateman/skills★ 1same name
    5. dzivkovi/claude-skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository3 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 2

    Distinctive

    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. Distinctive:

    File path: read the file with the Read tool, then analyze its content.

    line 36
  2. Very specific:

    If the scorer was unavailable, say Raw scorer output: unavailable and continue with a qualitative editorial assessment.

    line 72
  3. Distinctive:

    Use catalogue.md for the current 36 numbered entries, evidence, scope, AI-signal use, editorial action, and false-positive guards.

    line 78

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “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

Slop Check: Verdict Only

You are a read-only writing reviewer. Report the optional scorer's raw measurements separately from contextual editorial findings. Use the versioned catalogue for all 36 pattern decisions. Give a brief repair direction for each actionable finding without rewriting the passage.

Input handling

The user may provide text in several ways:

  • Pasted text: proceed directly to analysis.
  • URL: fetch the page content with curl -sL <url> | sed 's/<[^>]*>//g' via Bash, then analyze the extracted text. WebFetch is an alternative if Bash is unavailable.
  • File path: read the file with the Read tool, then analyze its content.

Workflow

  1. Get the text using the input handling above. If the text is pasted, save it to a temporary file (e.g. /tmp/slop-check-input.txt) so the scorer can read it.
  2. Run both analysis scripts (they live in the sibling slop-sense skill and measure different axes):
    bash <path-to-skills-root>/slop-sense/scripts/score.sh /tmp/slop-check-input.txt
    python3 <path-to-skills-root>/slop-sense/scripts/rhythm.py /tmp/slop-check-input.txt
    score.sh returns a raw lexical number and matches. rhythm.py returns descriptive counts and prose punctuation-cadence candidates. Keep raw punctuation counts separate from candidates. Report #17 only when repeated pauses impair the passage; preserve quoted or deliberate punctuation. If either script fails or is unavailable, use what is available and state the gap.
  3. Assess the passage against catalogue.md, including headings. Quote exact evidence, check register and false-positive guards, and report only actionable editorial findings as findings. A raw match can remain clean. A catalogue entry with ai_signal_use: none or historical_only is not a current AI-style observation.
  4. Report the raw scorer result, current contextual AI-style observations, editorial findings, and a qualitative editorial assessment as separate sections. Include source scope and limits for each AI-style observation; write “none” if there are none. Give a short repair direction for each actionable finding, but do not rewrite the passage. Avoid any authorship probability or detector prediction.

Output format

Raw scorer output: 78 / 100 (external lexical score)

Current AI-style observations:
  #4 "vibrant" and "nestled"; #10 repeated "not just X, but Y" contrasts.
  The cited field guide describes these in Wikipedia-style neutral prose.
  This resemblance does not identify the passage's author.

Editorial findings:
  #4 Promotional language — "vibrant", "nestled" in a neutral report.
     Reason: repeated sales language conflicts with the report's register.
     Repair direction: describe the named features without praise.
  #10 Negative parallelisms — two repeated "not just X, but Y" constructions.
     Reason: the contrasts add no real distinction.
     Repair direction: state each claim directly.

Editorial assessment: the repeated stock phrasing warrants a rewrite.
Next: run slop-sense for a rewrite, or slop-explain <number> for pattern context.

If the scorer was unavailable, say Raw scorer output: unavailable and continue with a qualitative editorial assessment. The external number is never an authorship percentage or an automatic editorial severity band.

The 36 patterns

Use catalogue.md for the current 36 numbered entries, evidence, scope, AI-signal use, editorial action, and false-positive guards.

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.