Agent skill

Anti Slop

by rand in rand/cc-polymath

Comprehensive toolkit for detecting and eliminating "AI slop" - generic, low-quality AI-generated patterns in natural language, code, and design.

MITAuto-check passedWriting & Content

Install Anti Slop

skills CLI
$ npx skills add rand/cc-polymath --skill anti-slop -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install rand/cc-polymath anti-slop --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/rand/cc-polymath.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/anti-slop .claude/skills/anti-slop && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
anti-slop
GitHub stars
181
Token cost
~2.9k tokens
SKILL.md length
1,101 words
Files
8 (incl. scripts, references)
Skills in repo
26
Repo updated
First seen
Licence
MIT

At a glance

Comprehensive toolkit for detecting and eliminating "AI slop" - generic, low-quality AI-generated patterns in natural language, code, and design.

  • Works in 2 steps: Detect Slop → Clean Slop
  • Improving content quality
  • SKILL.md covers What is AI Slop?, When to Use This Skill, Core Workflow and Text Slop Detection & Cleanup, plus 8 more sections
  • Runs Python scripts from its folder; calls python

What it does

Anti Slop is an agent skill from rand/cc-polymath. Comprehensive toolkit for detecting and eliminating "AI slop" - generic, low-quality AI-generated patterns in natural language, code, and design. Use when reviewing or improving content quality, preventing generic AI patterns, cleaning up existing content, or enforcing quality standards in writing, code, or design work.

Its SKILL.md is about 2.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 9 other files, including scripts and reference files (for example `CLAUDE_MD_UPDATES.md`, `README.md` and `references/code-patterns.md`).

It sits in Writing & Content, covering Humanizing AI text. The repository describes itself as: Claude Code skills and workflows, optimized for context-efficiency and skill quality. Skills ranging from cloud infrastructure to design to advanced maths. The licence is MIT.

When your agent uses it

  • Improving content quality
  • Preventing generic AI patterns
  • Cleaning up existing content
  • Enforcing quality standards in writing

Example prompts

  • “AI slop”
  • “/anti-slop”

Requirements

  • Python 3

Workflow steps

2 steps, taken from the step headings in SKILL.md.

  1. Detect Slop
  2. Clean Slop

What it can do on your machine

Read from SKILL.md and the folder at commit baa2df1. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 2 files in scripts/ (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Anti Slop loads about 2.9k tokens when it runs, and up to ~9.3k if it reads all its reference files. Until then it costs about 83 tokens; SKILL.md has 1,101 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~83
When it runs · the whole SKILL.md, loaded when a task matches
~2.9k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~9.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from rand/cc-polymath at commit baa2df1, republished under its MIT licence (© rand). 1,101 words, ~2,939 tokens.

Download SKILL.mdSave it as .claude/skills/anti-slop/SKILL.md (or your agent's skills folder). This skill also uses 7 other files; get the full folder from GitHub.
name
anti-slop
description
Comprehensive toolkit for detecting and eliminating "AI slop" - generic, low-quality AI-generated patterns in natural language, code, and design. Use when reviewing or improving content quality, preventing generic AI patterns, cleaning up existing content, or enforcing quality standards in writing, code, or design work.

Anti-Slop Skill

Detect and eliminate generic AI-generated patterns ("slop") across natural language, code, and design.

What is AI Slop?

AI slop refers to telltale patterns that signal low-quality, generic AI-generated content:

  • Text: Overused phrases like "delve into," excessive buzzwords, meta-commentary
  • Code: Generic variable names, obvious comments, unnecessary abstraction
  • Design: Cookie-cutter layouts, generic gradients, overused visual patterns

This skill helps identify and remove these patterns to create authentic, high-quality content.

When to Use This Skill

Apply anti-slop techniques when:

  • Reviewing AI-generated content before delivery
  • Creating original content and want to avoid generic patterns
  • Cleaning up existing content that feels generic
  • Establishing quality standards for a project
  • User explicitly requests slop detection or cleanup
  • Content has telltale signs of generic AI generation

Core Workflow

1. Detect Slop

For text files:

bash
python scripts/detect_slop.py <file> [--verbose]

This analyzes text and provides:

  • Slop score (0-100, higher is worse)
  • Specific pattern findings
  • Actionable recommendations

Manual detection: Read the appropriate reference file for detailed patterns:

  • references/text-patterns.md - Natural language slop patterns
  • references/code-patterns.md - Programming slop patterns
  • references/design-patterns.md - Visual/UX design slop patterns
2. Clean Slop

Automated cleanup (text only):

bash
# Preview changes
python scripts/clean_slop.py <file>

# Apply changes (creates backup)
python scripts/clean_slop.py <file> --save

# Aggressive mode (may slightly change meaning)
python scripts/clean_slop.py <file> --save --aggressive

Manual cleanup: Apply strategies from the reference files based on detected patterns.

Text Slop Detection & Cleanup

High-Priority Targets

Remove immediately:

  • "delve into" → delete or replace with "examine"
  • "navigate the complexities" → "handle" or delete
  • "in today's fast-paced world" → delete
  • "it's important to note that" → delete
  • Meta-commentary about the document itself

Simplify wordy phrases:

  • "in order to" → "to"
  • "due to the fact that" → "because"
  • "has the ability to" → "can"

Replace buzzwords:

  • "leverage" → "use"
  • "synergistic" → "cooperative"
  • "paradigm shift" → "major change"
Quality Principles

Be direct:

  • Skip preambles and meta-commentary
  • Lead with the actual point
  • Cut transition words that don't add meaning

Be specific:

  • Replace generic terms with concrete examples
  • Name specific things instead of "items," "things," "data"
  • Use precise verbs instead of vague action words

Be authentic:

  • Vary sentence structure and length
  • Use active voice predominantly
  • Write in a voice appropriate to context, not corporate-generic

Code Slop Detection & Cleanup

High-Priority Targets

Rename generic variables:

  • data → name what data it represents
  • result → name what the result contains
  • temp → name what you're temporarily storing
  • item → name what kind of item

Remove obvious comments:

python
# Bad
# Create a user
user = User()

# Better - let code speak
user = User()

Simplify over-engineered code:

  • Remove unnecessary abstraction layers
  • Replace design patterns used without purpose
  • Simplify complex implementations of simple tasks

Improve function names:

  • handleData() → what are you doing with data?
  • processItems() → what processing specifically?
  • manageUsers() → what management action?
Quality Principles

Clarity over cleverness:

  • Write code that's easy to understand
  • Optimize only when profiling shows need
  • Prefer simple solutions to complex ones

Meaningful names:

  • Variable names should describe content
  • Function names should describe action + object
  • Class names should describe responsibility

Appropriate documentation:

  • Document why, not what
  • Skip documentation for self-evident code
  • Focus documentation on public APIs and complex logic

Design Slop Detection & Cleanup

High-Priority Targets

Visual slop:

  • Generic gradient backgrounds (purple/pink/cyan)
  • Overuse of glassmorphism or neumorphism
  • Floating 3D shapes without purpose
  • Every element using same design treatment

Layout slop:

  • Template-driven layouts ignoring content needs
  • Everything in cards regardless of content type
  • Excessive whitespace without hierarchy
  • Center-alignment of all elements

Copy slop:

  • "Empower your business" type headlines
  • Generic CTAs like "Get Started" without context
  • Buzzword-heavy descriptions
  • Stock photo aesthetics
Quality Principles

Content-first design:

  • Design around actual content needs
  • Create hierarchy based on importance
  • Let content determine layout, not templates

Intentional choices:

  • Every design decision should be justifiable
  • Use patterns because they serve users, not because they're trendy
  • Vary visual treatment based on element importance

Authentic voice:

  • Copy should reflect brand personality
  • Avoid generic marketing speak
  • Be specific about value proposition

Reference Files

Consult these comprehensive guides when working on specific domains:

  • text-patterns.md - Complete catalog of natural language slop patterns with detection rules and cleanup strategies

  • code-patterns.md - Programming antipatterns across languages with refactoring guidance

  • design-patterns.md - Visual and UX design slop patterns with improvement strategies

Each reference includes:

  • Pattern definitions and examples
  • Detection signals (high/medium confidence)
  • Context where patterns are acceptable
  • Specific cleanup strategies

Scripts

Show full SKILL.md (459 more words)Show less
detect_slop.py

Analyzes text files for AI slop patterns.

Usage:

bash
python scripts/detect_slop.py <file> [--verbose]

Output:

  • Overall slop score (0-100)
  • Category-specific findings
  • Line numbers and examples
  • Actionable recommendations

Scoring:

  • 0-20: Low slop (authentic writing)
  • 20-40: Moderate slop (some patterns)
  • 40-60: High slop (many patterns)
  • 60+: Severe slop (heavily generic)
clean_slop.py

Automatically removes common slop patterns from text files.

Usage:

bash
# Preview changes
python scripts/clean_slop.py <file>

# Save changes (creates backup)
python scripts/clean_slop.py <file> --save

# Save to different file
python scripts/clean_slop.py <file> --output clean_file.txt

# Aggressive mode
python scripts/clean_slop.py <file> --save --aggressive

What it cleans:

  • High-risk phrases
  • Wordy constructions
  • Meta-commentary
  • Excessive hedging
  • Buzzwords
  • Redundant qualifiers
  • Empty intensifiers

Safety:

  • Always creates .backup file when overwriting
  • Preview mode shows changes before applying
  • Preserves content meaning (non-aggressive mode)

Best Practices

Prevention Over Cure

When creating content:

  1. Write with specific audience in mind
  2. Use concrete examples over abstractions
  3. Lead with the point, skip preambles
  4. Choose words for precision, not impression
  5. Review before considering it complete
Context-Aware Cleanup

Not all patterns are always slop:

Acceptable contexts:

  • Academic writing may need more hedging
  • Legal documents require specific phrasing
  • Internal documentation can use shortcuts
  • Technical docs have domain-specific conventions

Always consider:

  • Who is the audience?
  • What is the purpose?
  • Does this pattern serve a function?
  • Is there a better alternative?
Iterative Improvement
  1. Detect - Run detection scripts or manual review
  2. Analyze - Understand which patterns are truly problems
  3. Clean - Apply automated cleanup where safe
  4. Review - Manually verify changes maintain meaning
  5. Refine - Fix remaining issues by hand
Quality Over Automation

The scripts are tools, not replacements for judgment:

  • Use automated detection to find candidates
  • Apply automated cleanup to obvious patterns
  • Manually review anything that changes meaning
  • Exercise discretion based on context

Integration Patterns

Code Review
bash
# Check files before committing
python scripts/detect_slop.py src/documentation.md --verbose

# Clean up automatically
python scripts/clean_slop.py src/documentation.md --save
Content Pipeline
  1. Create initial content
  2. Run slop detection
  3. Apply automated cleanup
  4. Manual review and refinement
  5. Final quality check
Standards Enforcement

Create project-specific thresholds:

  • Max acceptable slop score: 30
  • Required manual review for scores > 20
  • Auto-reject submissions with scores > 50

Limitations

Scripts only handle text:

  • Code slop detection is manual (use code-patterns.md)
  • Design slop detection is manual (use design-patterns.md)

Context sensitivity:

  • Scripts can't understand all contexts
  • Some "slop" may be appropriate in certain domains
  • Always review automated changes

Language coverage:

  • Detection patterns optimized for English
  • Code patterns focus on common languages (Python, JS, Java)
  • Design patterns are platform-agnostic

Common Scenarios

Scenario 1: Review AI-Generated Content
bash
# User asks: "Can you review this article for AI slop?"
1. Read references/text-patterns.md for patterns to watch
2. Run: python scripts/detect_slop.py article.txt --verbose
3. Review findings and apply manual cleanup
4. Optionally run: python scripts/clean_slop.py article.txt --save
5. Do final manual review of cleaned content
Scenario 2: Clean Up Codebase
bash
# User asks: "Help me clean up generic AI patterns in my code"
1. Read references/code-patterns.md
2. Review code files manually for patterns
3. Create list of generic names to rename
4. Refactor following principles in code-patterns.md
5. Remove obvious comments and over-abstractions
Scenario 3: Design Review
bash
# User asks: "Does this design look too generic?"
1. Read references/design-patterns.md
2. Check against high-confidence slop indicators
3. Identify specific issues (gradients, layouts, copy)
4. Provide specific recommendations from design-patterns.md
5. Suggest concrete alternatives
Scenario 4: Establish Quality Standards
bash
# User asks: "Help me create quality standards for our team"
1. Review all three reference files
2. Identify patterns most relevant to user's domain
3. Create project-specific guidelines
4. Set up detection scripts in development pipeline
5. Document acceptable exceptions

Tips for Success

For text cleanup:

  • Run detection first to understand scope
  • Use non-aggressive mode for important content
  • Always review automated changes
  • Focus on high-risk patterns first

For code cleanup:

  • Start with renaming generic variables
  • Remove obvious comments next
  • Refactor over-engineered code last
  • Test after each significant change

For design cleanup:

  • Audit visual elements against patterns
  • Prioritize structural issues over aesthetic ones
  • Ensure changes serve user needs
  • Maintain brand consistency

General principles:

  • Quality > uniformity
  • Context > rules
  • Clarity > cleverness
  • Specificity > generality

© rand, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 7 other files (scripts, references) in skills/anti-slop of rand/cc-polymath.

  • SKILL.md
  • CLAUDE_MD_UPDATES.md
  • README.md
  • references/code-patterns.md
  • references/design-patterns.md
  • references/text-patterns.md
  • scripts/clean_slop.py
  • scripts/detect_slop.py

Open the folder on GitHubat commit baa2df1

Compare with similar skills

Anti Slop next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Anti Slop compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Anti Slop this skillrand/cc-polymath181—~2.9kAutomated safety check: PassMIT
HumanizerAzure-Samples/interview-coach-agent-framework17338 repos~5.8kAutomated safety check: PassMIT
Avoid AI Writingconorbronsdon/avoid-ai-writing4.9k3 repos~8.1kAutomated safety check: PassMIT
User-Facing Text Cleanupguillaumemeyer/watermarks-remover24k—~3.5kAutomated safety check: PassMIT
Install Anti Sloptrycompai/crm11k1 repos~881Automated safety check: PassMIT
Stop SlopXe/site7328 repos~423Automated safety check: PassMIT

Similar skills

  • Humanizer

    Azure-Samples/interview-coach-agent-framework

    Official

    Remove signs of AI-generated writing from text. An agent skill from Azure-Samples/interview-coach-agent-framework.

    173 GitHub starsUsed in 38 repos~5.8k tokens
    Writing & ContentAuto-check passed
  • Avoid AI Writing

    conorbronsdon/avoid-ai-writing

    Audit and rewrite content to remove AI writing patterns ("AI-isms").

    4.9k GitHub starsUsed in 3 repos~8.1k tokens
    Writing & ContentAuto-check passed
  • User-Facing Text Cleanup

    guillaumemeyer/watermarks-remover

    Audits prose for invisible Unicode characters and rewrites it while keeping facts, citations, code and required disclosures unchanged and the writer's voice intact.

    24k GitHub stars~3.5k tokensUpdated yesterday
    Writing & ContentAuto-check passed
  • Install Anti Slop

    trycompai/crm

    Install and configure the anti-slop Oxlint plugin in a local TypeScript or JavaScript repository.

    11k GitHub starsUsed in 1 repo~881 tokens
    Writing & ContentAuto-check passed
  • Stop Slop

    Xe/site

    Remove AI writing patterns from prose. An agent skill from Xe/site.

    732 GitHub starsUsed in 8 repos~423 tokens
    Writing & ContentAuto-check passed
  • Chinese Text Humanizer

    op7418/Humanizer-zh

    Edits Chinese articles, comments and documents to remove filler, repetition and template phrasing while keeping the facts, the level of certainty and the author's voice.

    19k GitHub stars~2k tokensUpdated 17 days ago
    Writing & ContentAuto-check passed

More from rand/cc-polymath

All 26 skills in this repo
  • Typed Holes Refactor

    rand/cc-polymath

    Refactor codebases using Design by Typed Holes methodology - iterative, test-driven refactoring with formal hole resolution, constraint propagation, and continuous validation.

    181 GitHub stars~5.6k tokensUpdated 7 mo ago
    Auto-check passed
  • Elegant Design

    rand/cc-polymath

    Create world-class, accessible, responsive interfaces with sophisticated interactive elements including chat, terminals, code display, and streaming content.

    181 GitHub stars~2.7k tokensUpdated 7 mo ago
    Auto-check passed
  • Discover Agentic

    rand/cc-polymath

    Automatically discover agentic workflow skills when building AI agents, implementing tool use patterns, managing context windows, decomposing complex tasks, or designing multi-step autonomous…

    181 GitHub stars~1.4k tokensUpdated 7 mo ago
    Auto-check passed
  • Discover API

    rand/cc-polymath

    Automatically discover API design skills when working with REST APIs, GraphQL schemas, API authentication, OAuth, JWT, rate limiting, API versioning, error handling, or endpoint design.

    181 GitHub stars~1.5k tokensUpdated 7 mo ago
    Auto-check passed
  • Discover Database

    rand/cc-polymath

    Automatically discover database skills when working with SQL, PostgreSQL, MongoDB, Redis, database schema design, query optimization, migrations, connection pooling, ORMs, or database selection.

    181 GitHub stars~2k tokensUpdated 7 mo ago
    Auto-check passed
  • Discover Frontend

    rand/cc-polymath

    Automatically discover frontend development skills when working with React, Next.js, UI components, state management, data fetching, forms, accessibility, performance optimization, or SEO.

    181 GitHub stars~2.4k tokensUpdated 7 mo ago
    Auto-check passed

Questions about Anti Slop

What does Anti Slop do?

Comprehensive toolkit for detecting and eliminating "AI slop" - generic, low-quality AI-generated patterns in natural language, code, and design. Anti Slop is an agent skill from rand/cc-polymath. Comprehensive toolkit for detecting and eliminating "AI slop" - generic, low-quality AI-generated patterns in natural language, code, and design.

When should I use Anti Slop?

Anti Slop fits situations like: improving content quality; preventing generic AI patterns; cleaning up existing content; enforcing quality standards in writing.

How do I install Anti Slop in Claude Code?

Run `npx skills add rand/cc-polymath --skill anti-slop -a claude-code`. Or copy the skill folder (skills/anti-slop in rand/cc-polymath) into .claude/skills/anti-slop in your project. Claude Code loads it when a task matches its description.

How do I install Anti Slop in Codex?

Run `npx skills add rand/cc-polymath --skill anti-slop -a codex`. Or copy the skill folder (skills/anti-slop in rand/cc-polymath) into .agents/skills/anti-slop in your project. Codex loads it when a task matches its description.

Can I use Anti Slop in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add rand/cc-polymath --skill anti-slop -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/anti-slop, .gemini/skills/anti-slop, .github/skills/anti-slop and .opencode/skills/anti-slop in your project.

What does Anti Slop need to run?

Going by SKILL.md and its folder, Anti Slop needs Python for the scripts in its folder and the command-line tools its instructions call (python). Our summary lists: Python 3.

Does Anti Slop access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Anti Slop safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Anti Slop use?

Anti Slop is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Anti Slop use?

About 2.9k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 6.4k tokens, read only when the agent opens those files.

What are the alternatives to Anti Slop?

Skills that share tags, products or a category with Anti Slop: Humanizer (Azure-Samples/interview-coach-agent-framework, 173 stars), Avoid AI Writing (conorbronsdon/avoid-ai-writing, 4.9k stars), User-Facing Text Cleanup (guillaumemeyer/watermarks-remover, 24k stars) and Install Anti Slop (trycompai/crm, 11k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Anti Slop?

rand (a GitHub user) maintains it in rand/cc-polymath, which has 181 GitHub stars. The repository holds 26 skills in this directory. The repository was last updated on February 28, 2026.

Source: rand/cc-polymath on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.