Evo MCP Server now available
Snyk's Evo tool now integrates with MCP clients like Claude, Codex, and Cursor, allowing users to query their AI assets, visualize connections, review policies, and manage policy violations directly from their agent. Users can discover AI assets across code, developer machines, and pentest targets without leaving their workflow.
You can connect any MCP client, such as Claude, Codex or Cursor, to Evo and ask about your AI estate from the agent you already work in.
What's new. You can now:
Ask what AI assets Evo has discovered across your code, your developer machines, and your pentest targets
See how those assets connect, for example which MCP servers an agent uses or which findings came from a target
Read your policies and the violations open against them
Create and update policies without leaving your agent
To get started, check out our documentation
Source: original entry ↗
More from Snyk
Follow Snyk to get its new changes in your feed and email digest.
Snyk Code October update: improved analysis and more findings
Snyk Code now analyzes projects more deeply and accurately across Java, Kotlin, Go, TypeScript, JavaScript, Python, C#, PHP and Ruby, with significantly more findings reported. Improvements include better import resolution via config files, corrected PHP typing, expanded library coverage, new vulnerability detection rules, and enhanced template and serverless application analysis.
Snyk CLI v1.1308.0 released
Introducing HTML report generation for test commands, experimental agent-scan improvements with risk indicators, and substantial reliability fixes for Maven, Gradle, and .NET projects to prevent stack overflows and scanning failures.
ADS installer now available as packaged binary
The Agent Supply Chain Security installer now ships as signed macOS .pkg and Windows .msi packages, eliminating the need for custom scripts and MDM integration. The installer registers scheduled jobs automatically for independent scan execution.