Every protection layer, every notification type, every detection source — documented here so you know exactly how MailShield Guardian keeps you safe.
Take a photo of any letter, envelope, or postcard. AI reads the full text using OCR and detects fraud patterns in seconds.
| Step | What Happens |
|---|---|
| 1. Take Photo | Camera captures the letter or envelope — no uploading to a server manually. The app handles it automatically. |
| 2. OCR Extraction | Google Cloud Vision reads every word on the page — sender name, addresses, phone numbers, URLs, account numbers. |
| 3. AI Analysis | Claude AI classifies the content against known scam patterns: fake government agencies, prize notices, Medicare fraud, overdue bill scams, and more. |
| 4. Community Cross-Check | The sender's phone number or address is checked against our community database of reported scammers. |
| 5. Result Delivered | SAFE / SUSPICIOUS / DANGER — with a plain-English explanation of exactly what the threat is and what to do. |
Paste any suspicious message — text, email body, or forwarded content — and get an instant fraud risk assessment with a plain-English explanation.
| Message Type | What We Detect |
|---|---|
| SMS / Text | Delivery scams, bank impersonation, gift card requests, "click here" phishing links, prize notifications, fake two-factor codes. |
| Phishing, spoofed sender addresses, fake invoice requests, CEO impersonation, account compromise notices, subscription renewal fraud. | |
| Forwarded Messages | Chain messages, "share this warning" scams, fake news designed to harvest clicks or personal information. |
Scan any QR code before you tap it. Quishing (QR phishing) is the fastest-growing fraud vector — fake QR codes on parking meters, restaurant tables, and mail are used to steal credentials and payment info. MailShield checks the URL hidden inside before you go anywhere near it.
| Step | What Happens |
|---|---|
| 1. Point Camera | Open MailShield and tap "Check a QR Code." Point your camera at any QR code — on mail, a sign, a receipt, a menu, or a screen. |
| 2. URL Extracted | The URL encoded in the QR code is captured instantly — without you navigating to it. |
| 3. AI Analysis | Claude AI analyzes the domain, path, and URL structure for phishing patterns, credential-harvesting redirectors, and known malicious domains. |
| 4. Community Cross-Check | The domain is checked against our community database of reported phishing sites and malicious URLs. |
| 5. Result Delivered | SAFE / SUSPICIOUS / DANGER — before your browser ever loads the page. If it's dangerous, close the app and ignore the QR code entirely. |
Known scammers are blocked before your phone even rings. They can't call through, and they can't leave a voicemail. Completely silent protection.
| Platform | How It Works | Voicemail Blocked? |
|---|---|---|
| 📱 iOS | Integrated with Apple's native call screening system — blocked callers hear "number not available" before the call ever connects. | ✅ Yes — caller never connects |
| 🤖 Android | Integrated with Android's native call screening system — blocked calls are rejected at the OS level with a busy signal. | ✅ Yes — busy signal prevents most carrier voicemails |
| Scenario | What Happens | Notification Fired |
|---|---|---|
| Known scammer calls (cached) | Instantly blocked before the phone rings. No voicemail possible. | 🛡️ "Scam Call Blocked — No voicemail delivered" |
| Unknown number — AI flags as HIGH RISK | Call is allowed through on first contact (AI check runs in real time). Result is cached. Next call from this number is instantly blocked. | ⚠️ "Scam Call — Will Be Blocked — cached for next time" |
| Unknown number — AI flags as SUSPICIOUS | Call is allowed through. Number is labeled with a warning and monitored. | ⚠️ "Suspicious Call from [number]" |
| Community database match | Blocked immediately — same as a cached scammer. Community-verified numbers are distributed to all users automatically. | 🛡️ "Scam Call Blocked — Community Database" |
Before you send money, gift cards, wire transfers, or Zelle payments — verify the request here first. Payment fraud is irreversible. This screen is the last line of defense.
| Payment Type | Common Scams Detected |
|---|---|
| Gift Cards | "Send iTunes gift card codes" — IRS, tech support, grandchild scam. Gift cards are untraceable and unrecoverable. |
| Wire Transfer | Romance scam wires, fake contractor deposits, business email compromise (BEC) redirects. |
| Zelle / Venmo / Cash App | Facebook Marketplace fraud, fake invoices, overpayment scams, "accidental" payment reversals. |
| Cryptocurrency | Investment fraud, "double your money" schemes, fake exchange wallets, romance crypto scams. |
| Check / Money Order | Fake check scams, overpayment fraud, mystery shopper schemes. |
Turn it on once and forget about it. Auto-Protect monitors your incoming texts, emails, and calls in the background — 24 hours a day, without you having to do anything.
| Channel | How Auto-Protect Works |
|---|---|
| 📱 SMS Texts | Monitors incoming texts in the background. Flagged messages trigger an instant push notification before you even open the conversation. |
| 📧 Gmail | Connects via Gmail API (read-only). Scans incoming email subjects and senders. Suspicious emails are flagged before you see them. |
| 📧 Outlook | Connects via Outlook/Microsoft API (read-only). Same background scanning as Gmail. |
| 📧 IMAP / Other Email | Connect any email provider via IMAP. Scans inbox for threats on a regular interval. |
| 📞 Phone Calls | Every incoming call is screened. Known scammers are blocked instantly. New numbers are checked in the background and cached for next time. |
Watch over parents, grandparents, or any family member remotely. Get a real-time notification the moment a threat is detected on their device — labeled with their name so you always know who needs help.
| Scenario | What the Family Member Sees |
|---|---|
| Parent opens scam text | Auto-Protect catches it and pushes: "🤖 Auto-Protect: 🚨 DANGER — SMS · Mom: IRS scam text detected. Be cautious." |
| Parent receives scam email | Push notification: "Mom: sender@irs-refunds.net · IRS Refund Phishing" |
| Parent scans suspicious mail | Family sees the scan result in their Family Dashboard — labeled "Mom's Scan" with the risk level and explanation. |
| Parent's scam call is blocked | Family alert: "🛡️ Auto-Protect blocked a call for Mom from (602) 555-0134 — known scammer." |
Our proprietary fraud intelligence network — built from multiple trusted sources and continuously growing with every MailShield Guardian user's experience. When one person catches a scam, everyone benefits.
An animated live map showing verified scam reports near you — updated in real time as the community submits new reports. See what threats are active in your area right now, color-coded by risk level.
| Map Element | What It Means |
|---|---|
| 🔴 Red dot | High-risk verified scam report — active threat in that area |
| 🟠 Orange dot | Suspicious report — possibly a scam, community flagged as worth knowing |
| 🟢 Green dot | Community report marked safe — useful context for tracking false alarms in your area |
| 📍 My Area button | Instantly zooms the map to your local ZIP area with one tap |
| Tap any dot | View the scam type, brief description, and when it was reported |
A live, searchable feed of community scam reports in your area. Read what your neighbors have encountered, filter by scam type, engage with reactions and comments, and help others stay safe — all anonymously.
| Feature | How It Works |
|---|---|
| 🔍 Search | Search the feed by keyword — "Medicare," "gift card," "IRS," "fake check," etc. Results update as you type. |
| 🏷️ Type Filter | Tap any category chip to filter: IRS, Bank Fraud, Package, Romance, Crypto, Tech Support, Medicare, Gift Card, Other. Tap again to clear. |
| ❤ Like | Like a report to upvote it and make it more visible to others in the area. |
| 😰 Reactions | Add an emoji reaction — Helpful, Scary, Happened to Me, or Spreading — to give the community more nuanced feedback. |
| 💬 Comments | Comment on any report (up to 280 chars) to add context, share your own experience, or ask questions. Completely anonymous. |
| 🚩 Flag | Flag inappropriate reports. After 3 flags, the report is auto-hidden pending moderation review. |
| 🔔 Comment Notifications | Get a push notification when someone comments on a report you submitted — so you can stay in the conversation. |
Share a scam you almost fell for — anonymously. No name, no email, no judgment. Just your experience helping someone else avoid the same trap.
Protecting yourself and your community earns recognition. Badges mark real milestones in your fraud awareness journey. Your Fraud Resilience Score — a number from 0 to 850 — measures your overall engagement and grows every time you take action.
| Score Component | How Points Are Earned |
|---|---|
| 📬 Mail Scans | Points per physical mail scan completed |
| 💬 Text/Email/Call Scans | Points per manual or Auto-Protect scan |
| 📢 Community Reports | Higher points for submitting verified scam reports to BlockBrief |
| 🤫 Confession Booth Posts | Points for sharing near-miss stories anonymously |
| 💬 Comments | Points for contributing helpful comments on community reports |
| 👍 Reactions | Points for engaging with community reports via reactions |
| ✅ Feedback Accuracy | Bonus points for marking scan results accurately (correct/false positive/missed) |
When a scan returns a high-risk result, MailShield automatically surfaces one-tap contact buttons for the four major US fraud reporting agencies — so help is always one tap away when you need it most.
| Agency | What They Do | Best For |
|---|---|---|
| 🏛️ FTC — ReportFraud.ftc.gov | Federal Trade Commission consumer fraud reporting | Any consumer scam, identity theft, robocalls |
| 🔍 FBI IC3 | Internet Crime Complaint Center — cybercrime reporting | Online fraud, phishing, business email compromise |
| 📞 AARP Fraud Watch Network | Free helpline at 1-877-908-3360 — volunteer fraud specialists | Phone scams, elder fraud, in-person guidance |
| 🏦 CFPB | Consumer Financial Protection Bureau — financial fraud | Bank fraud, wire transfers, financial institution complaints |
Stay informed about local scam activity even when you haven't opened the app. The Weekly Area Digest is an optional push notification sent once per week summarizing the top threats reported near your ZIP code.
Every notification MailShield Guardian can send — organized by category, with the exact message text and trigger condition.
| Notification | Trigger | Source Tag |
|---|---|---|
| 🚨 "Scam Text Detected" | AI classifies an SMS as HIGH risk | 🤖 Claude AI or ⚡ Auto-Protect |
| ⚠️ "Suspicious Text" | AI classifies an SMS as SUSPICIOUS | 🤖 Claude AI or ⚡ Auto-Protect |
| 🚨 "Scam Email Detected" | AI classifies an email as HIGH risk | ⚡ Auto-Protect |
| ⚠️ "Suspicious Email" | AI classifies an email as SUSPICIOUS | ⚡ Auto-Protect |
| 🚨 "Scam Mail Detected" | Mail photo scan returns HIGH risk | 🤖 Claude AI + 🌐 Community DB |
| ⚠️ "Suspicious Mail" | Mail photo scan returns SUSPICIOUS | 🤖 Claude AI |
| ⚠️ "Scam Call Detected" | AI flags an incoming call as HIGH risk (call allowed, result cached) | 🤖 Claude AI |
| ⚠️ "Suspicious Call" | AI flags an incoming call as SUSPICIOUS | 🤖 Claude AI |
| 🛡️ "Scam Call Blocked" | Cached scammer attempts to call — blocked instantly before phone rings | ⚡ Auto-Protect or 🌐 Community DB |
| ⚠️ "Scam Call — Will Be Blocked" | New number flagged after async AI check — now cached for instant blocking next time | 🤖 Claude AI |
| 🤖 "Auto-Protect: DANGER — [type]" | Auto-Protect catches a scam; pushed to family members with member label | ⚡ Auto-Protect |
| ⚠️ "Auto-Protect: SUSPICIOUS — [type]" | Auto-Protect catches a suspicious item; pushed to family | ⚡ Auto-Protect |
| 🔗 "Cross-Channel Alert" | Same scammer detected across multiple channels (text + call, or call + email) | 🔗 Cross-Channel |
| 💳 "Payment Risk Detected" | Payment verification check returns HIGH or SUSPICIOUS risk | 🤖 Claude AI |
Every result shows a detection badge so you always know what intelligence flagged it. No black box — complete transparency about why something was marked as a threat.
| Detection Badge | What It Means |
|---|---|
| 🌐 Community DB | Matched against our community database of confirmed scam numbers, domains, and patterns — built from trusted sources and verified across the MailShield user network. |
| 🔗 Cross-Channel | The same threat actor was spotted across multiple channels on your device — for example, a phone number that appeared in both a scam text and an incoming call. Elevated risk automatically. |
| 🤖 Claude AI | Analyzed in real time by Claude AI — evaluating message content, sender patterns, urgency tactics, impersonation signals, and known fraud signatures in full context. |
Every user gets unlimited scanning for free. Pro upgrades the AI engine, removes throttles, and adds monitoring & family features.
| Feature | Free | ⭐ Pro |
|---|---|---|
| Scan physical mail, texts, calls, email, QR & payments | ✅ | ✅ |
| AI analysis engine | Haiku → Sonnet escalation on text/SMS/email Haiku only on physical mail |
Haiku → Sonnet escalation on text/SMS/email Sonnet directly on physical mail |
| Scan quantity | Unlimited | Unlimited |
| Multi-mail batch scanning (up to 5 at once) | 1 at a time | Up to 5 at once |
| Auto-Protect background scanning (Gmail / Outlook / IMAP / SMS / Call) | ❌ | Real-time, always on |
| Gmail & Outlook inbox monitoring | ❌ | ✅ |
| Call blocking & community database | ✅ | ✅ |
| Family Dashboard & real-time alerts | ❌ | ✅ |
| Cross-channel correlation (same scammer across SMS, email & calls) | ❌ | ✅ |
| Scan history view | Last 3 only | Unlimited |
| Community & Social Features | ||
| 🗺️ Live Scam Radar map | ✅ | ✅ |
| 📖 BlockBrief community feed (read, search, filter) | ✅ | ✅ |
| ❤ Likes, reactions & comments on reports | ✅ | ✅ |
| 💬 Confession Booth (anonymous near-miss posts) | ✅ | ✅ |
| 🆘 Emergency contacts on high-risk results | ✅ | ✅ |
| 📬 Weekly Area Digest notification | ✅ | ✅ |
| 🏅 Community Defender Badges | ✅ | ✅ |
| 📊 Fraud Resilience Score (0–850) | ✅ (visible to all) | ✅ + 25 pt Pro boost |
| 👤 Profile screen (badges, score, reports) | ✅ | ✅ |
| 🎙️ Voicemail Scanner | ✅ | ✅ |
| 👵 Senior Mode (larger text, simpler flow) | ✅ | ✅ |
| 🤝 Trusted Contacts (skip scans for senders you've marked safe) | ✅ | ✅ |
| 🧭 Recovery Wizard (steps to take after a scam) | ✅ | ✅ |
| 📡 Regional Scam Feed | ✅ | ✅ |
| Price | $0 forever | $6.99/mo or $49.99/yr |
Pro Annual saves 49% vs monthly. All Pro plans include a 7-day free trial.
Family plan — $9.99/mo or $89.99/yr — adds the shared Family Dashboard on top of everything in Pro, for up to 5 members via Apple Family Sharing.
Free unlimited scanning. No account required. Upgrade for full protection.
⬇ Download on the App Store ⬇ Get it on Google Play