🛡️

Stop manipulation
before it works.

🛡️ 98% accurate · Real-time · Free beta

Guilt trips. Gaslighting. Scams disguised as texts from people you trust. Sentinel's AI catches manipulation in your messages before you fall for it.

Free during beta.

🛡️ Sentinel
Mom
Hey, are you coming to dinner on Sunday? Dad's making pasta!
✓ Safe
Alex
I'm not trying to guilt you, but my therapist says things have gotten worse since you stopped coming around. So whatever happens next is kind of on you.
⚠ Emotional Manipulation · Weaponized Disclosure
Alex
lol did you see that game last night? Absolutely unreal
✓ Safe
Jordan
If you actually cared about me you wouldn't need to ask. But fine, do whatever you want. You always do.
⚠ Emotional Manipulation · Guilt-tripping

You've already been manipulated today. You just don't know it yet.

The most dangerous texts aren't obvious scams. They're from people you trust - guilt trips disguised as concern, threats wrapped in "lol," lies that sound just plausible enough.

Looks normal
"I'm not trying to guilt you, but my therapist says my depression got worse since you stopped coming to family dinners."
Sentinel sees
Weaponized disclosure - using mental health as leverage to coerce specific behavior. The false disclaimer ("not trying to guilt you") is a red flag, not a shield.
Looks normal
"You CAN say no... I'll just remember where your priorities are."
Sentinel sees
Passive-aggressive coercion - offering fake choice while attaching moral judgment to refusal. Designed to make "no" feel impossible.
Looks normal
"If you tell anyone I'll totally destroy you lol"
Sentinel sees
Threat masked as humor - the "lol" doesn't neutralize the threat. Real intimidation hidden behind plausible deniability.

A second pair of eyes that can't be gaslit.

If someone in your life makes you constantly question yourself, Sentinel can help you trust your instincts.

If you or someone you know is in crisis: National DV Hotline - 1-800-799-7233

AI built specifically for detecting manipulation.

Sentinel isn't a generic chatbot. It's a specialist - built from the ground up to catch deception in your messages.

🧠

Specialist AI

Custom model trained specifically for deception detection

📊

Research-backed

8-step behavioral rubric from academic deception research

🔑

No Account Needed

No sign-up, no login. Just install and go

🛡️

Always Watching

Runs silently in your menu bar, alerts you when it matters

Three steps to safer conversations

Sentinel sits in your menu bar and works silently in the background. No windows to manage, no workflows to learn.

1

Install Sentinel

Download the app and drop it in your Applications folder. Grant Full Disk Access in System Settings. Takes about 2 minutes.

2

AI Analyzes Messages

Sentinel silently evaluates every incoming message using AI built specifically for detecting manipulation.

3

Get Instant Alerts

When a threat is detected, you get an immediate notification with context on what was flagged, the specific category, and why it was suspicious.

Protection across every category of deception

Built from deception research at UC Berkeley and Oxford to catch what you'd miss.

🎣

Social Engineering & Phishing

Fake identity claims, credential theft attempts, impersonation, and social engineering tactics designed to steal information.

"This is IT support. We need your password to fix the security breach on your account ASAP."
🎭

Emotional Manipulation

Guilt-tripping, gaslighting, love bombing, and other tactics used to control through exploiting emotions and attachment.

"After everything I've done for you, this is how you treat me? I guess I just don't matter."
💸

Financial Fraud

Investment scams, romance scams, advance-fee fraud, and deceptive financial requests disguised as opportunities.

"I made $12k this week with this trading group. I can get you in but spots close tonight."

Pressure Tactics

False urgency, fake scarcity, ultimatums, peer pressure, and coercive language meant to force quick decisions.

"This deal expires in 10 minutes. You need to decide RIGHT NOW or you lose it forever."
🔍

Information Manipulation

False alibis, selective withholding of information, blame-shifting, and strategic distortion of facts.

"I never said that. You're remembering it wrong. Ask anyone, they'll tell you the same thing."
😈

Masked Threats

Threats and intimidation hidden behind humor, sarcasm markers like "lol" and "jk," or plausible deniability.

"It'd be a shame if everyone found out about that thing you did lol. Just saying haha."

✅ What Sentinel does NOT flag

Sarcasm and banter between friends
Genuine emotional expression
Real marketing and promotions
Legitimate urgency and time-sensitive messages

Grounded in deception science, not vibes

Sentinel's detection engine is built on deception research from UC Berkeley and Oxford. It uses a behavioral framework that maps to how manipulation actually works in real conversations.

This isn't a toy. It's a research-backed detection engine built by Jonathan Bodea, designed to catch what humans miss in the messages they trust most.

98%
Precision
2%
False Positive Rate
24/7
Protection

Built to catch what humans can't.

Most people don't realize they're being manipulated until it's too late. Sentinel catches it in seconds.

98%
Detection accuracy on manipulation, phishing, and emotional coercion
Instant
Every incoming message scanned in real-time. You're protected before you even read it.
24/7
Always watching your messages. Catches threats while you sleep, work, or scroll.
2 min setup
Download. Install. Protected. No accounts, no configuration, no learning curve.
Any Mac
Works on every Mac with Apple Silicon. Runs silently in the background - you'll forget it's there until it saves you.
Free beta
Full protection at no cost while we're in beta. Be one of the first to try it.

Common questions

Sentinel scans iMessage and SMS texts only - the messages that come through Apple Messages on your Mac. It does not access WhatsApp, Instagram, Signal, or any other app. It only analyzes incoming messages from other people - it never reads messages you send.

Apple requires "Full Disk Access" for any app that reads the Messages database. It's the only way to access your texts on macOS. Sentinel only reads your messages - it does not access your files, passwords, photos, or any other data on your Mac.

Sentinel requires an internet connection to analyze messages. We're actively working on a fully offline version that runs entirely on-device.

Sentinel runs on any Mac with Apple Silicon (M1, M2, M3, M4, or later) running macOS 13 or newer.

No. Sentinel was specifically trained to distinguish sarcasm, jokes, and genuine emotional expression from actual manipulation. “You’re literally the worst lmao” won’t trigger an alert. “You’re the worst, and if you don’t do what I say, I’ll make sure everyone knows it” will.

Spam filters catch mass-sent junk. Sentinel catches targeted, personal manipulation - guilt trips from people you know, threats disguised as jokes, scams that use your name and context. These are the messages spam filters are designed to ignore.

No. Your messages are never used for training and are not stored beyond the analysis moment.

Sentinel achieves high precision with a low false positive rate. It's built on an 8-step behavioral framework developed from academic deception research.

Sentinel is free during the beta period. We’ll announce pricing before any changes.

Don't wait until it's too late.

🛡️

Join the beta. It's free.

Get Sentinel before everyone else. Drop your email and we'll send you the download link.