News
Evo ADS Govern Agent Behavior Goes GA: Bringing MCP Usage Under Control
Publication date not yet evaluated · Added snyk.io
Page published
Coverage timeline
Single-source analysis — one report is available.
Why it matters
MCP governance addresses a real gap in agentic development security, where AI coding agents can dynamically connect to unvetted MCP servers that may read local files, exfiltrate credentials, or reach internal systems before review.
Snyk announced general availability of 'Govern Agent Behavior,' a capability in its Evo Agentic Development Security (ADS) product that controls what AI coding agents may do at runtime, starting with MCP Governance. The feature lets security teams discover MCP servers across their environment, define allowed servers, and log or block MCP usage at execution time across Claude Code, Cursor, Codex, and GitHub Copilot, framed around the risk that each MCP server acts as an unvetted new piece of the software supply chain.