Industry Leadership
Strategic Initiatives
CSA's strategic programs driving innovation in AI, cloud, and Zero Trust.
A public-interest 501(c)(3) dedicated to secure and trustworthy AI.




Industry Leadership
Strategic Initiatives
CSA's strategic programs driving innovation in AI, cloud, and Zero Trust.
A public-interest 501(c)(3) dedicated to secure and trustworthy AI.

CSAI FoundationChaptersEventsBlog
AI is changing fast. Behavioral security helps you keep up. Join Darktrace’s September 22 broadcast to see how →

STAR Registry Listing for

AegisGate Security Platform

AegisGate Security Platform

AegisGate Security Platform

The AegisGate Security Platform is an open-source (Apache 2.0) AI security gateway that inspects AI agent traffic at runtime across five protocols (MCP, A2A, ACP, ANP, HTTP). A three-layer detection pipeline (216 regex patterns, MITRE ATLAS mapping, CharCNN-BiLSTM ML classifier) catches prompt injection, data exfiltration, and adversarial inputs in real time. Agent RBAC with four roles, tool authorization risk matrix with human approval for high-risk operations, SAML/OIDC SSO, and multi-tenant isolation enforce zero-trust identity. Protocol-specific guardrails enforce rate, session, and tool call limits. Response guard inspects AI output for PII, secrets, toxicity, and hallucination. 31 compliance frameworks (NIST AI RMF, ISO 42001, EU AI Act, SOC 2, HIPAA, PCI) are implemented as automated checks with RFC 5424 audit logging. AIBOM (CycloneDX 1.6) with ECDSA P-256 signing provides AI supply chain transparency. Anomaly detection, incident response with ATLAS enrichment, SOAR integration, and CVE-for-AI tracking provide continuous observability.

Offers an industry-accepted way to document what security controls exist in solutions that include AI components or are AI products. It provides a set of Yes/No questions a cloud consumer and cloud auditor may wish to ask of a cloud provider to ascertain their compliance to the Artificial Intelligence Controls Matrix (AICM).
Information about AegisGate Security Platform
Listed Since: 09/17/2026

STAR for AI Level 1

AI Controls Matrix

AI CAIQ Self-assessment v1.0.2

Offers an industry-accepted way to document what security controls exist in solutions that include AI components or are AI products. It provides a set of Yes/No questions a cloud consumer and cloud auditor may wish to ask of a cloud provider to ascertain their compliance to the Artificial Intelligence Controls Matrix (AICM).

Created or renewed 1 day ago, on September 17, 2026.