API Usage
INPUT TOKENS0
OUTPUT TOKENS0
ESTIMATED COST$0.0000

Based on Claude Sonnet 4.6 pricing ($3 / 1M input tokens, $15 / 1M output tokens). Tracks usage from this browser only — does not include other devices or direct Anthropic console usage. The counter is shared across all laboratory tools. Reset clears this browser's count only.

// tool 01 · rhetoric analysis

Fnord Detector

Paste any text — a sales page, AI feedback, email, article — and find out what persuasion techniques are quietly doing the work. Great for checking whether your AI assistant is actually giving you honest feedback, or just telling you what you want to hear. Identifies manipulation patterns, scores density, and strips the text to show you what's really there.

Works best on 100–2000 words. Longer texts take a moment.

// SCANNING ...
CATEGORY KEY
fearTriggers anxiety or threat perception authorityAppeals to credentials to bypass critical thought identityUs vs them framing, tribal signals urgencyArtificial time pressure, FOMO emotionalCharged language designed to override reason assumptionDebatable claims presented as obvious facts missing contextSelective framing that distorts reality
FEAR
HYPE
URGE
FLAT
// AWAITING TEXT INPUT
Paste text on the left — sales copy, AI feedback, an email —
and see exactly what persuasion techniques are at work.