The centralized platform to discover and secure all your AI agents, enabling safer enterprise adoption. We work in Agentic Security: the action layer where AI agents call tools, run multi-step workflows and talk to other agents is a new attack surface that filtering at the model boundary alone cannot cover.
| TrustGate | Open-source AI gateway for LLM and agent traffic. |
| TrustGuard for Cursor | AI firewall for the Cursor agent. |
| AI-Gateway-Benchmark | Reproducible performance comparison vs Kong, Tyk, KrakenD, Apache APISIX. |
| echo-chamber | Research code for the Echo Chamber jailbreak (paper). |
Our research has been adopted into the OWASP GenAI Security taxonomy:
- GenAI Red Teaming Guide
- OWASP GenAI Data Security Risks and Mitigations initiative (2026 report)
- [Vendor Evaluation Criteria for AI Red Teaming Providers & Tooling](https://genai.owasp.org/resource/owasp-vendor-evaluation-criteria-for-ai-red- teaming-providers-tooling-v1-0/)
- Leader (Product & Innovation Leadership), [2025 KuppingerCole Leadership Compass for Generative AI Defense](https://neuraltrust.ai/guides/2025-kuppingercole-leadership-compass-for- generative-ai-defense)
- Included in Gartner Market Guides
- Selected for the EIC Accelerator by the European Innovation Council
Slack · @neuraltrust · Blog
