Civily.ai Chrome extension security report

Civily.ai is a parental control tool that filters YouTube comments in real-time using AI. Every comment is hidden instantly and only revealed after it passes moderation — your child never sees toxic content, even briefly. HOW IT WORKS • Comments are hidden the moment they appear on screen • A keyword filter catches obvious violations instantly • Remaining comments are assessed by Claude AI for nuanced content moderation • Only comments that pass both checks are revealed • Any error defaults to hiding — safety first 10 MODERATION CATEGORIES Toggle each category on or off from the settings popup: • Profanity & Slurs • Sexual Content • Violence • Substance Use • Harassment & Targeted Abuse • Dog-Whistle Racism • Identity-Based Hate • Self-Harm & Suicide Content • Predatory Contact Attempts • Disinformation (opt-in, high threshold) PARENT CONTROLS • Password-protected settings popup • Per-category on/off toggles • Adjustable sensitivity (relaxed / standard / strict) • Real-time block statistics (today / this week / all time) • Shield status indicator (green / yellow / red) PRIVACY BY DESIGN • Comment text is processed in memory and never stored • Only metadata (scores, hashed identifiers) retained for 90 days • No browsing history collected • Works only on YouTube — no other sites • Full privacy policy: https://api.civily.io/privacy REQUIREMENTS • Active subscription ($2.99/month via Stripe) • Works on YouTube desktop (youtube.com and youtu.be) in Google Chrome • English language comments only SECURITY • All API communication over HTTPS • Auth tokens bound to specific installations • Backend holds all API keys — nothing sensitive in the extension • Fail-safe: any error hides all comments (shield-down mode)