Verity

AI Output Verification & Trust Scoring
MODE: HEURISTIC SCAN
v0.3 — pattern-based, no model call
Verity flags writing patterns like
  • Absolute words ("always," "100%")
  • Numbers with no source attached
  • Vague backing ("studies show...")
  • Overconfident certainty
  • Cause-and-effect claims with no proof
It cannot tell you
  • Whether something is actually true
  • Real-world facts, people, or events
  • Anything needing outside knowledge
Try an example:
Restore your last scanned text?
0 characters
—style score
—
—
This is a writing-style score, not a fact-check. A high score means the phrasing sounds careful — it doesn't mean every claim inside is correct. Always double-check specifics yourself.
Scan ID: — Analyzed in — Scan confidence: —
Overall Assessment
—
—
0
Critical flags
0
Risky flags
0
Caution flags
0
Sentences scanned
Critical Risky Caution
Findings