Skill

slop-radar

AI slop detection skill for Claude Code. Scans text for AI buzzwords, structural patterns, and scoring.

Usage

When reviewing or writing text, check it for AI slop patterns. Run slop-radar on any text file, clipboard content, or inline text to get a quality score.

Instructions

When the user asks to check text for AI slop, buzzwords, or writing quality:

1. Read the target text (file, selection, or inline)
2. Scan for these categories:

Buzzword detection (English)

Check for these high-frequency AI phrases (partial list -- see full database in slop-radar repo):

• Filler: "dive into", "delve", "landscape", "journey", "tapestry", "realm", "endeavor"
• Hype: "game-changer", "cutting-edge", "revolutionary", "transformative", "unprecedented", "groundbreaking"
• Corporate: "leverage", "synergy", "stakeholder", "bandwidth", "paradigm shift", "value-add"
• Connectors: "moreover", "furthermore", "additionally", "it's worth noting", "needless to say"
• Openers: "in today's world", "as we navigate", "let me break this down", "here's the thing"
• Actions: "streamline", "optimize", "harness", "unlock", "empower", "catalyze", "reimagine"

Buzzword detection (German)

• "bahnbrechend", "wegweisend", "transformativ", "ganzheitlich", "massgeschneidert"
• "darueber hinaus", "nichtsdestotrotz", "in anbetracht", "zusammenfassend"
• "potenzial entfalten", "impulse setzen", "massstaebe setzen"

Structural patterns

• Excessive em-dash usage (3+ per paragraph)
• "Let me [verb]" sentence starters
• "Here's the thing/what" openers
• Binary contrasts ("Not just X, but Y")
• Wh- question headers ("What makes this", "Why this matters")
• Bullet-point overload (6+ consecutive bullets)
• Triple adjective chains
• Meta-references ("In this article/post/guide")
• High passive voice density (>30% of sentences)
• Emoji + bold header combinations

Scoring

Start at 100, deduct:

• -2 per buzzword found
• -1 to -5 per structural pattern (by weight)
• -10 for high passive voice density
• -3 per "Let me" / "Here's" opener

Bonus +5 for questions, +5 for varied sentence length.

Rating: 90-100 HUMAN, 70-89 MOSTLY CLEAN, 50-69 SUSPICIOUS, 30-49 LIKELY AI, 0-29 PURE SLOP

Output format

Report the score, list found buzzwords, list detected patterns, and suggest concrete replacements:

• Replace "leverage" with "use"
• Replace "utilize" with "use"
• Replace "comprehensive" with "complete" or "full"
• Replace "robust" with "solid" or "strong"
• Replace "seamless" with "smooth"
• Replace "innovative" with describe what is actually new
• Replace "transformative" with describe the actual change
• Drop filler connectors entirely ("moreover", "furthermore", "additionally")
• Replace "it's worth noting" with just state the thing
• Replace "in today's fast-paced world" with be specific about what changed

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: renefichtmueller
• Source: renefichtmueller/slop-radar
• License: MIT
• Homepage: https://slop-radar-demo.pages.dev

Install and usage instructions live in the source repository linked above.