Deception & Attribution Platform
Make attackers reveal themselves.
Deploy intelligent decoys across your network, your endpoints, and your customer-facing AI. When threats interact with them, you learn who, what, and where.
AI Deception · v1.0.5
Turn an LLM jailbreak attempt into a labeled threat signal.
Squirrelops sits in front of your customer-facing language model. When someone tries to jailbreak it, exfiltrate secrets, or prompt-inject your agent, we route them into a high-fidelity decoy and capture exactly what they tried, what tooling they used, and what they were after.
Your real model stays clean. Your security team gets a feed of labeled attacker behavior.
Internal adversarial campaign, v1.0.4 (2026-05-12). Methodology: see the AI page.
Products
Three lines, one platform.
AI · New
Deception for the
LLM attack surface.
Sit in front of your customer-facing AI. Route jailbreak, prompt-injection, and credential-probing traffic into a high-fidelity decoy. Feed labeled attacker behavior straight to your SIEM.
Enterprise
Stop chasing breaches.
Start controlling them.
Full deception security platform for organizations. Deploy decoy environments, profile attackers, and generate structured threat intelligence, all from a single operator dashboard.
Home · Free
Know your network.
Control your network.
Native macOS app for home network monitoring. Discover every device, deploy invisible tripwires, and get alerts you can actually trust: legitimate devices never touch a decoy.
Community Edition is on the roadmap.
The current product repositories are private. We are working toward a self-contained, open-source Community Edition. The roadmap is proposed, and no release date is committed.
Community Edition overviewPilot engagement
Run an evaluation pilot.
We work with security teams that have a defined LLM attack surface and an internal red-team or pen-test capability. A 4-to-6-week engagement with a signed profile bundle scoped to your model and use cases.
Request a pilot