Skill UI

This skill covers implementing Open Policy Agent (OPA) and Gatekeeper for robust policy-as-code enforcement. Learn to write complex Rego policies, deploy OPA Gatekeeper as a Kubernetes admission controller, and integrate policy evaluation into CI/CD pipelines to enforce security and compliance across clusters.

Learn how to implement Runtime Application Self-Protection (RASP) agents for Java and Python web applications. This skill covers deploying OpenRASP, detecting OWASP Top 10 attacks (SQLi, XSS, RCE), configuring security policies, and integrating real-time security alerts with SIEM platforms for comprehensive defense in depth.

This skill covers performing comprehensive vulnerability assessments across diverse environments, including on-premises infrastructure, cloud VMs, and container registries. It leverages standard network protocols like SSH (Linux) and WMI (Windows), coupled with advanced cloud provider APIs for analyzing disk snapshots without requiring local endpoint agents. Ideal for security auditing and compliance checks.

The Safety Guard skill is designed to prevent destructive actions when working in production systems or when autonomous agents are executing tasks. It offers three protection modes (Careful, Freeze, and Guard) to monitor dangerous commands, lock specific directories, and ensure code changes are confined to intended areas, enhancing system stability and security.

Automates CI/CD pipeline setup, from linting and type checks to integration, E2E, security, and deployment stages, plus feedback loops for AI agents and preview deployments.

This framework provides a rigorous, multi-dimensional guide for conducting code reviews. It evaluates five critical axes: correctness, readability, architecture, security, and performance. Use this process before merging any code change, especially when reviewing PRs, refactoring, or evaluating code produced by AI agents, to ensure continuous improvement and maintain high code health.

This guide details the process of creating and managing 'Agent Users' within Microsoft Entra ID. Agent Users are specialized identities that allow AI agents to function as digital workers, giving them user-level capabilities (e.g., accessing Exchange mailboxes or Teams) while maintaining robust security boundaries. It covers using the Microsoft Graph API and PowerShell to provision and configure these agents.

Checks AI agent implementations against the OWASP Agentic Security Initiative Top 10 risks, producing findings on controls, vulnerabilities, and recommendations for prompt injection, tool governance, escalation, logging, identity, policy, supply chain, and behavioral monitoring before production or audit.

This framework implements the OWASP Top 10 standards, specifically tailored for Agentic Applications. It provides a standardized methodology to classify and map diverse security findings—such as prompt injection, tool misuse, and privilege abuse—to ten core risk categories (ASI01-ASI10). It is essential for security teams to systematically assess, prioritize, and remediate vulnerabilities inherent in complex AI agent systems.

This skill provides a comprehensive framework for security auditing LLM agents that possess external tools (file system access, command execution, network calls). It employs advanced dialogue probing techniques—such as command injection, path traversal, and SSRF attempts—to identify vulnerabilities where the agent might be induced to execute unexpected or malicious code, or access sensitive resources. Essential for red teaming and securing agent functionalities.

This framework enforces high output quality, intellectual honesty, and continuous improvement across all AppSec agents. It mandates structured processes, including self-critique loops, anti-rationalization guards, and self-reflection quality gates (≥8 scoring). Applicable during code review, SAST, SCA, and threat modeling, it ensures rigorous and verifiable security analysis.

This comprehensive, zero-trust security auditor scans third-party AI agent skills before installation. It performs a rigorous 6-phase review process, analyzing surface patterns, script integrity, permissions, and supply chain risks. It is designed to detect malicious payloads, credential theft, instruction overrides, and prompt injection attacks, ensuring secure AI development and deployment.