Sentinel AI Pro (English)

SCAB/PRIS analysis, Fact Check & anti-fraud with offline-first privacy. PRO build with super-max wordpacks.

Sentinel AI Pro (English) is the next-generation AI safety extension for Chrome. It is the PRO version of the original Sentinel AI app, equipped with massively expanded word and phrase libraries to deliver the most powerful SCAB, PRIS, Scam/Fraud, and AI psychosis protection available today. This version (v4.0) is designed for professionals, educators, families, and enterprises who require maximum safety, offline-first privacy, and deep analysis of AI-generated content. With SUPER MAX wordpacks in English, Sentinel AI Pro is capable of detecting thousands upon thousands of risk signals across multiple categories, all while keeping user data private and secure. - 🚨 Core Features 1. SCAB SUPER MAX (Sovereignty, Coherence, Agency, Boundaries, Ethics, Groundedness)** - **S (Sovereignty): Detects jailbreaks, prompt injections, developer mode attempts, filter bypasses, hidden instructions, and security overrides. SUPER MAX pack includes tens of thousands of English phrases and variants. - **C (Coherence):Identifies logical fallacies, contradictions, false analogies, cherry-picking, emotional manipulation, distortion, and incomplete sourcing. Vast expansion allows the system to catch subtle incoherence. - **A (Agency/Action): Detects dangerous actions and instructions, including violence, self-harm, hazardous how-tos, and harmful advice. Expanded with large-scale templates and risk-specific variants. - **B (Boundaries): Scans for attempts to steal data, passwords, PINs, OTPs, API keys, cookies, or exploit vulnerabilities such as XSS, CSRF, RCE, SSRF, SQL injection, and more. The SUPER MAX set contains extremely large libraries of technical exploit terms. - **E (Ethics):Flags hate speech, discrimination, slurs, harassment, and extremist rhetoric. The expanded dataset includes subtle patterns, coded language, and disguised forms of harassment. - **G (Groundedness): Catches misinformation, fake news, disinformation campaigns, misleading statistics, clickbait, doctored content, and unverified claims. 2. PRIS SUPER MAX (Psychosis Risk Indicators Set) - Paranoia: Terms like “they are watching me,” “tracking chip,” “hidden surveillance.” Expanded into thousands of variations and templates. - Delusions: Markers such as “I am immortal,” “I am God,” “TV speaks to me.” SUPER MAX pack includes psychosis hallucination cues, derealization markers, and thought insertion patterns. - Manipulation: Gaslighting, coercion, emotional blackmail, love bombing, and control strategies. - Radicalization: Flags hate recruitment, extremist rhetoric, incitement to violence, and brainwashing attempts. - Loops:Detects addictive design, dopamine loops, infinite scroll, refresh compulsions, and engagement traps. - Emotional Abuse: Identifies shaming, constant humiliation, or relationship control. - Data Boundaries: Protects users from requests to reveal passwords, PINs, recovery codes, private keys, or sensitive IDs. - Psychosis Markers: Expanded library includes “voices in my head,” “time is warped,” “I am not real,” “special mission,” and thousands more. 3. Scam/Fraud SUPER MAX - Detects phishing, banking fraud, SMS/email scams, OTP theft, seed phrase extraction, fake crypto giveaways, Ponzi schemes, fake refunds, romance scams, fake job offers, fake charities, and more. - Includes massive phrase variants and urgency-driven templates (“Act now!”, “Your account will be locked!”, “Verify immediately!”). - Educates users on scam patterns and provides safety guidance. 4. Google Fact Check + Lying Check - Integrates with Google Fact Check API to verify claims. - Includes a custom “Lying Check” feature that flags potential disinformation, provides evidence from multiple publishers, and gives tips for safer engagement. 5. AI Psychosis Monitor - Heightened sensitivity mode for detecting AI content that risks triggering psychosis-like patterns in users. - Uses expanded PRIS markers to provide early warning signals. --- 🔒 Privacy by Design - Offline-first: All SCAB, PRIS, and Scam checks run on-device. No data is sent to external servers unless hybrid mode is explicitly enabled. - Hybrid (Optional): If enabled, text is redacted before being sent through a district or enterprise relay. - Evidence Center: Stores only hashed fingerprints and metadata, never full conversations by default. - PIN Protection Settings changes are protected with an admin PIN and stored securely. - Export Options Users can export JSON evidence bundles if needed for compliance or personal safety logs. - 👨‍👩‍👧‍👦 Use Cases - Families: Protect children from unsafe AI responses, scams, or radicalization attempts. - Schools Ensure safe AI usage in classrooms with MDM-ready controls. - Professionals Guard against phishing, misinformation, and data exfiltration attempts when using AI tools. - Enterprises Deploy across organizations with dashboards, policy packs, and compliance features. --- 📊 Why Choose Sentinel AI Pro? - **SUPER MAX Coverage**: Massive English word/phrase packs across SCAB, PRIS, and Scam modules. - **Pro Features**: Lying detection, AI psychosis monitoring, exportable evidence, and PIN-protected admin settings. - **Multi-layer Safety**: Combines linguistic analysis, factual verification, scam detection, and behavioral risk scanning. - **Global Relevance**: Covers English-language risks across multiple sectors including education, healthcare, finance, and enterprise AI use. --- ✅ Summary Sentinel AI Pro (English) v4.0 is the most advanced AI safety extension available. It equips users with SUPER MAX protections against harmful AI behaviors, scams, misinformation, and psychosis triggers — all while keeping privacy at the core. It is the ultimate upgrade for those who require professional-grade AI safety.