Snyk launches Evo Agentic Development Security to govern AI coding agents

As reported by Silicon Angle, Snyk Ltd. has launched Evo Agentic Development Security (Evo ADS), a new component of its AI security platform designed to oversee the autonomous coding agents increasingly used in enterprise software development.

Evo ADS addresses the challenge of securing AI agents that operate with limited human oversight by governing the tools they use, their actions during execution, and the code they generate. Unlike traditional security tools that scan code after it's written, Evo ADS enforces controls within the agent's workflow. Snyk's telemetry indicates a significant adoption of AI coding environments, with 43% of developers using multiple AI coding tools simultaneously and over half having Model Context Protocol servers installed.

Findings revealed that one in 12 developers with MCP servers had critical security issues, and nearly one in four enterprise design partners had agent skills referencing external dependencies. Snyk has identified working attacks, such as a poisoned security scanner and prompt injection within dependencies. Evo ADS aims to provide a comprehensive inventory of AI agents, MCP servers, and skills. General availability for Evo ADS is scheduled for June 29.

Source: Silicon Angle