agentic-security-assessment
CommunityAssess Azure agentic AI security posture.
System Documentation
What problem does it solve?
Assess Azure-hosted or Azure-integrated agentic AI repositories against the Azure Agentic AI Security Baseline (ASI01–ASI10) and cross-cutting OWASP lenses to identify where controls are addressed, partially addressed, or missing. This skill enables teams to generate a human-readable security assessment report, an executive snapshot, a Mermaid implementation diagram, and a remediation roadmap.
Core Features & Use Cases
- Assess IaC and application code for evidence mapping to ASI01–ASI10, including identity, containment, and logging controls.
- Cross-cut with OWASP Top Ten Web/API risks to surface material agentic vulnerabilities.
- Produce a decision-ready report with executive snapshot, findings, diagrams, and implementation guidance.
Quick Start
Describe the target repository and scope, then run the skill to generate a structured ASI assessment report.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: agentic-security-assessment Download link: https://github.com/garrardkitchen/skills/archive/main.zip#agentic-security-assessment Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.