Official open-source developer SDKs and Digital Forensics & Incident Response (DFIR) tools for Sealed Rose — AI Deepfake Detection, Synthetic Media Analysis, and Frame-Level Forensics.
Built for DFIR analysts, SOC teams, threat intelligence researchers, and incident response pipelines investigating:
- Executive & Employee Impersonation (synthetic video/audio in business communications)
- Identity & KYC Fraud (face swap manipulations in onboarding videos)
- Social Engineering & Phishing (generative media campaigns)
- Digital Evidence Verification (frame-by-frame anomaly detection and model attribution)
DFIR teams can instantly triage suspicious media files without an account or API setup using the free web verification engine: 👉 Sealed Rose Deepfake Video & Image Detector
| Language / Tool | Package / Image | Documentation |
|---|---|---|
| JavaScript / TypeScript | npm install sealedrose |
npm Package · SDK README |
| Python | pip install sealedrose |
PyPI Package · Python README |
| PHP | composer require sealedrose/sealedrose-sdk |
Packagist Package · PHP README |
| Ruby | gem install sealedrose |
RubyGems Package · Ruby README |
| Docker | docker pull derekgallardo01/sealed-rose-detector |
Docker Hub · Docker README |
| VS Code Extension | Sealed Rose Media Detector |
Visual Studio Marketplace |
Ideal for incident response scripts, Jupyter forensic notebooks, and automated IR triaging pipelines:
pip install sealedrosefrom sealedrose import SealedRose
client = SealedRose()
# Triage a suspicious video file collected during an incident
result = client.verify_video(file_path="evidence/suspicious_exec_call.mp4")
print(f"Verdict: {result.get('verdict')}") # e.g., "AI Deepfake Detected"
print(f"Confidence Score: {result.get('deepfakeScore')}") # 0.0 to 1.0
print(f"Forensic Indicators: {result.get('indicators')}")npm install sealedroseimport { SealedRose } from 'sealedrose';
const client = new SealedRose();
// Verify video authenticity via public URL or local stream
const result = await client.verifyVideo({
url: 'https://incident-evidence.internal/case-2026-092/payload.mp4'
});
console.log(result.verdict); // "Looks Real" or "AI Deepfake Detected"
console.log(result.deepfakeScore); // Confidence ratingFor air-gapped sandboxes or containerized security pipelines:
docker run -it --rm derekgallardo01/sealed-rose-detector:latest- Frame-Level Artifact Detection: Analyzes compression discrepancies, optical flow discontinuities, facial landmark micro-jitters, and frequency-domain anomalies.
- Generative AI Attribution: Identifies artifacts characteristic of diffusion engines (Midjourney, Stable Diffusion, Flux, Sora, Runway Gen-3, Kling, Google Veo).
- Fast Triaging API: Designed to plug directly into SOAR workflows, SIEM alerting hooks, and ticketing pipelines.
- Web Verification: Sealed Rose Video Forensics
- Image & Face Authenticity: Sealed Rose Image Forensics
- Chrome Extension: Sealed Rose on Chrome Web Store
- Firefox Add-on: Sealed Rose on Firefox Add-ons
This project and its SDKs are licensed under the MIT License — see the LICENSE file for details.