# TrueTraceShorts — Digital Red Flags Strategy Update

Session learning: when producing TrueTraceShorts / AutoShortsBot content, the official primary positioning is no longer AI workflow tips, prompt engineering, or abstract productivity mechanisms.

## Official positioning

**TrueTraceShorts = Digital Red Flags / Scam Self-Defense for Normal People**

Core promise:

> One screen. One red flag. One safer move.

Audience-facing promise:

> I show you the one red flag before you click, pay, scan, or log in.

German internal explanation:

> Ein normaler Screen. Ein gefährlicher Hinweis. Eine sichere Handlung.

## Audience

Do not frame this externally as a “boomer channel”. Internally, content may be boomer-safe: simple language, big readable text, concrete examples, no jargon, clear action.

Target viewers:

- normal smartphone/internet users
- parents and teenagers
- working people
- small business owners
- families and teams
- people who want to protect parents, partners, colleagues, or friends

Psychological test:

> The viewer should think: “That could have happened to me or my family.”

## Language rules

Final social videos may remain English, but use simple B1/B2 everyday English:

- short sentences
- no corporate security language
- no nerd jargon unless immediately explained
- no buzzwords

Good:

- “This message wants you to hurry.”
- “Don’t tap the link.”
- “Open the app yourself.”
- “If the payment route changed, verify it another way.”
- “If it makes you hurry, slow down.”

Bad:

- “This phishing attempt leverages urgency-based social engineering to exploit cognitive load.”
- “Validate the authenticity of the communication channel before initiating a transactional flow.”

Plain replacements:

- “validate communication channel” → “check it another way”
- “credential harvesting” → “stealing your login”
- “MFA prompt fatigue” → “a login prompt you did not start”

## Default content formula

Every candidate must answer:

1. What normal screen does the viewer see?
2. What concrete red flag is visible?
3. Is the problem visible in second 1 without sound?
4. What could happen to the viewer?
5. What is the one safer move?
6. Can someone send this to a parent, partner, colleague, or team?
7. Is there a save/screenshot reason?
8. Is the language useful without tech knowledge?

Default structure:

- 0–2s: visible trap + hook
- 2–7s: why it feels normal
- 7–15s: one red flag
- 15–24s: safer move
- 24–30s: screenshot checklist
- 30–34s: memorable rule + soft CTA

Do not render if the red flag and safer move are not clear.

## Official content pillars

Primary pillars:

1. Text & Delivery Scams — package SMS, customs fee, toll fee, short links, fake tracking.
2. Login & Account Traps — fake bank/Microsoft/Google/Apple login, fake reset, account-closure threats, login prompt not started by user.
3. Payment & Invoice Red Flags — fake invoice, changed bank route, CEO payment request, QR payment trap.
4. Marketplace & Buyer Scams — off-platform buyer, fake shipping link, escrow, overpayment, courier pickup.
5. Family & Emotion Scams — “Hi mom new number”, emergency money, grandparent scam, wrong-number scam, simple AI voice warning.
6. Fake Support / Popup Scams — infected-device popup, call support now, fake antivirus, browser lock, remote access request.
7. Password / Account Safety Basics — password reuse, password manager basics, MFA basics, recovery email risk.
8. Job / Side-Hustle Scams — fake recruiter, task scam, pay to get paid, remote job deposit, crypto onboarding trap.

Secondary/specialist only:

- AI workflow automation
- prompt engineering
- tool reviews
- top AI tools
- abstract productivity mechanisms
- cybersecurity news
- hacker tactics
- OSINT
- VPN comparisons
- dark-web explainers
- technical attack instructions

## Official series logic

1. One Screen. One Red Flag.
2. Send This To Your Parents
3. Before You Click
4. Money Move Red Flags
5. Account Trap

## Visual rules

Default style: screen/mockup-led.

Prefer:

- phone screen
- SMS
- email
- invoice
- chat
- pop-up
- login
- QR situation
- marketplace chat
- simple fictional mockups

Avoid as primary visual layer:

- abstract glass clouds
- generic laptop B-roll
- pretty but empty AI imagery
- funnel/ring metaphors
- PowerPoint-like diagrams
- unnecessary Wan/I2V/T2V

Scam screens should usually be deterministic renderer-owned, not freely invented by image generation. This preserves:

- no real brands/logos
- no real bank data, domains, phone numbers, QR codes, or invoices
- no dangerous operational templates
- clear mobile readability

## AI image standard

OpenAI/Codex image generation with `gpt-image-2-high` may be used for premium styleframes/backgrounds, but for scam content the safe default is:

```text
AI layer = text-free atmosphere/background/keyframes
Deterministic renderer = scam screen, UI, invoice/email/SMS/chat text, captions, labels, claims
```

AI images are acceptable when they provide atmosphere only and contain no important text/data. Do not ask AI to freely invent fake invoices, bank pages, SMS, or login pages.

QA generated backgrounds before use. Reject backgrounds with pseudo-text, logos, QR/barcodes, people/hands, bank-like details, or busy composition that competes with captions.

## Screenshot checklist rules

End-card/checklist should usually exist:

- max 3 points
- very short words
- large font
- 1.5–3 seconds visible
- mobile-safe
- title like “SAVE THIS” or “Quick check”

Examples:

```text
SAVE THIS

Delivery text?
Check:

Small fee?
Short deadline?
Suspicious link?

Open the app yourself.
```

## Scoring priorities

Prioritize:

- viewer pain in first second
- screen clarity without audio
- red flag clarity
- safe move clarity
- family/share potential
- screenshot checklist value
- beginner language score
- mockup safety score
- no-fearmongering score
- pretty-empty-visual blocker

Block a candidate if it looks beautiful but does not show a concrete problem, uses high-jargon language, has unclear red flag, lacks a safer next step, or the scene is generic enough to fit 50 unrelated videos.

## First prioritized candidate sequence after fake-invoice proof short

1. `delivery-sms-trap-35s`
   - Hook: “This delivery text is designed to make you panic.”
   - Takeaway: “If it makes you hurry, slow down.”

2. `fake-support-popup-call-now-30s`
   - Hook: “If a website tells you to call support, stop.”
   - Takeaway: “Real support does not hijack your screen.”

3. `marketplace-buyer-off-platform-30s`
   - Hook: “The buyer is not buying. They are moving you off-platform.”
   - Takeaway: “Keep the chat and payment inside the platform.”

4. `fake-qr-bank-login-page-35s`
   - Hook: “The QR code was not the problem. The page after it was.”
   - Takeaway: “If money or login is involved, open the app yourself.”

5. `password-reuse-domino-30s`
   - Hook: “One leaked password does not stay on one website.”
   - Takeaway: “One account, one password.”

## Delivery SMS render lesson

For `delivery-sms-trap-35s`, the better version used high-quality AI backgrounds only as atmosphere and deterministic foreground UI overlays:

- generated 3 portrait backgrounds with `gpt-image-2-high`
- used 2 backgrounds
- discarded 1 because a package label had barcode-like marks
- deterministic SMS used fictional `example.com`
- final MP4 was explicitly verified as `1080x1920`, `SAR 1:1`, `DAR 9:16`

If edge-tts word boundary events are missing, manually derive caption timings from sentence timings and render active-word ASS captions; do not drop subtitles silently.