Skip to content

KryptosAI/mcp-observatory

v1.30.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 17d MCP Developer Tools
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agent-security ai-agent ai-security ai-supply-chain cli code-scanning
+14 more
developer-tools github-action github-code-scanning mcp mcp-ci mcp-security mcp-server mcp-testing model-context-protocol regression-testing sarif schema-drift security supply-chain-security

Summary

AI summary

Adds skill scanning for malicious behaviors and Unicode obfuscation detection.

Full changelog

What's New

  • 🕵️ Skill scanning: detect credential theft, exfiltration, remote execution, prompt injection in agent skill files
  • 🌐 13+ agent auto-discovery: Cursor, Windsurf, VS Code, OpenCode, Codex, Gemini CLI, and more
  • 🔤 Unicode obfuscation detection: zero-width characters, bidi overrides, Unicode tag sequences
  • 🧬 YARA-inspired compound rules: credential-exfiltration, supply-chain-hijack, hidden-execution, social-engineering
  • 📊 572 tests, 59 test files

New Contributors

  • @albatrossflyon-coder made their first contribution in https://github.com/KryptosAI/mcp-observatory/pull/201

Full Changelog: https://github.com/KryptosAI/mcp-observatory/compare/v1.29.2...v1.30.0

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track KryptosAI/mcp-observatory

Get notified when new releases ship.

Sign up free

About KryptosAI/mcp-observatory

Regression testing for MCP servers. Auto-discovers servers from Claude configs, checks capabilities, invokes tools, detects schema drift between versions, and recommends new servers based on your environment. Works as both a CLI and an MCP server.

All releases →

Related context

Beta — feedback welcome: [email protected]