EmbedIDS: Open-source intrusion detection for embedded systems and physical AI

Lightweight, open-source runtime security. On-device anomaly detection for low-power embedded and IoT devices, with no dynamic memory allocation.

View on GitHub

Apache 2.0

C11 static library

No heap allocation

Works offline

demo-edge-001 Jetson-class device · simulated Pause

GPU hijack Model tampering Data exfiltration Telemetry loss

Device trustVerified

Complete TutorialComprehensive learning guideExamplesReady-to-run code samplesCase Study: Runtime Security for Edge AIEmbedIDS protecting a Jetson-class AI device

EmbedIDS is a lightweight Runtime Intrusion Detection SDK specifically designed for low-power embedded Internet of Things (IoT) devices. Our open-source SDK provides AI-driven protection to address the critical security gap in constrained IoT environments.

EmbedIDS provides on-device, AI-driven protection tailored to resource-constrained embedded systems. By deploying compact anomaly-detection agents directly on devices, we enable real-time protection without relying on external connectivity or cloud dependencies.

Our free SDK empowers developers to secure embedded devices at no cost, featuring:

The IoT security landscape faces unique challenges:

EmbedIDS addresses these challenges by providing:

Start securing your embedded devices immediately with our open-source SDK:

EmbedIDS thrives on open-source collaboration and community contribution:

Ready to secure your embedded devices? Download our open-source SDK and join the community!

Read the Quick Start

Edit page Last updated: Oct 14, 2026

Next Quick Start

Linuscent © 2026 Linuscent Ltd Apache 2.0 Latest documentation Report a docs issue Cookie settings Help improve the documentation

With your permission, we use Google Analytics to understand which guides are useful. You can change your choice at any time.