Skill Doctor - The Advanced AI Agent Security Engine

Skill Doctor GitHub Action is now live!

You can now integrate the industry's most advanced, multi-layered risk mitigation engine directly into your CI/CD pipelines to automatically scan AI agent skills before they execute.

Core Capabilities

• Layer 1 (Static Analysis): Blazing-fast pattern matching using YARA-X and Tree-sitter AST parsing.
• Layer 2 (Semantic Analysis): Intent extraction and contextual evaluation via an OpenAI-compatible REST API.
• Layer 3 (Behavioral Sandbox): Runtime execution tracing in a secure microsandbox environment to detect live exfiltration or command injection.
• Layer 4 (Threat Intelligence): Cryptographic hash matching against distributed, known-malicious community threat datasets.

How to use it

Drop this step into any workflow to automatically scan your repository's AI skills and fail the build if high-severity threats are detected:

steps:
  - uses: KalarisLabs/Skill-Doctor@main
    with:
      target: '.'
      fail-on: 'HIGH'


**Full Changelog**: https://github.com/KalarisLabs/Skill-Doctor/compare/v0.2.3...v1.0.0

**Full Changelog**: https://github.com/KalarisLabs/Skill-Doctor/compare/v0.2.3...v1.0.0

**Full Changelog**: https://github.com/KalarisLabs/Skill-Doctor/compare/v0.2.3...v1.0.0

**Full Changelog**: https://github.com/KalarisLabs/Skill-Doctor/compare/v0.2.3...v1.0.0

v0.2.3

Full Changelog: v0.2.2...v0.2.3

Full Changelog: v0.2.2...v0.2.3

Full Changelog: v0.2.2...v0.2.3

Full Changelog: v0.2.2...v0.2.3

Full Changelog: v0.2.2...v0.2.3

v0.2.2

Full Changelog: v0.2.1...v0.2.2