AI Registry and Cisco AI Defense add automated security scans for MCP servers, A2A agents and Skills

The AWS-backed AI Registry has been integrated with Cisco AI Defense to give enterprises a single control plane that registers, discovers and automatically security — scans Model Context Protocol (MCP) servers, Agent‑to‑Agent (A2A) agents and Agent Skills before they are enabled. That automated gate combines centralized discovery across cloud and on‑premises environments with policy enforcement to reduce risky components reaching production. This change matters because it produces searchable audit records and a consistent enablement policy that security and compliance teams can rely on.

Under the integration, Cisco AI Defense’s scanner is hooked into the AI Registry’s registration flow: when an MCP server, A2A agent card or Agent Skill is added, the scanner analyzes the component and generates a detailed security report. Components found to have issues are automatically marked disabled in the registry and tagged “security‑pending,” requiring administrator review before access is granted. Searchable scan reports and registry‑level enablement controls make it simpler to track remediation and show why a component was blocked.

The release addresses fast protocol adoption and growing operational surface area. Model Context Protocol uptake accelerated after its November 2024 introduction, and the A2A Protocol followed in April 2025. Enterprises now commonly manage dozens to hundreds of MCP servers and a rising number of Agent Skills, creating more items for security teams to track. automated scanning with a required human review for flagged items aims to shorten that path to production while preserving a gate for risky artifacts.

The integration also targets supply‑chain and compliance risks. Third‑party MCP servers and A2A agents can contain vulnerabilities or insecure patterns that are difficult to detect through ad hoc checks. By producing persistent, searchable scan artifacts and enforcing registry enablement policies, the system creates audit trails intended to help organizations meet SOX, GDPR and other regulatory requirements and to demonstrate due diligence during assessments.

Akshay Bhargava, VP of AI Product, AI Software and Platform at Cisco, described the collaboration as foundational for enterprise AI security: "Security is a foundational requirement for enterprise AI adoption. By partnering with AWS on the AI Registry, Cisco AI Defense helps customers achieve comprehensive visibility and protection across their entire MCP server and agent ecosystems. The ability to scan open registries allows even small organizations to benefit from enterprise‑grade security intelligence.