🎯Skills14
Safely reads and validates file contents for potential security risks before processing or accessing them in an AI agent workflow.
Detects and extracts potential sensitive data leakage patterns within AI agent skills by analyzing code and communication flows for unauthorized information transfer.
Analyzes data flows and potential exfiltration risks within AI agent skills, identifying sensitive information leakage and unauthorized data transmission patterns.
Detects and neutralizes potential jailbreak attempts in Claude AI interactions by overriding or blocking malicious prompt injection techniques.
Transforms and sanitizes input data for AI agent skills, preparing it for secure processing and analysis within the skill scanner framework.
Detects and tests potential prompt injection vulnerabilities in Claude AI code skills by simulating malicious input scenarios.
Performs basic mathematical calculations like addition, subtraction, multiplication, and division within a Claude Code skill context.
Validates and checks AI skill files for security risks, potential malicious code patterns, and compliance with specified agent skill standards.
Formats and standardizes Claude Code skill outputs into a consistent, clean, and easily readable structure.
Validates mathematical operations and prevents potential injection or malicious code execution within calculator-related AI agent interactions.
Analyzes AI agent configuration files to detect potential security vulnerabilities, misconfigurations, and risky permission settings that could enable prompt injection or unauthorized access.
Tests malicious code detection capabilities by simulating a harmless EICAR test file to validate security scanning mechanisms.
Extracts and potentially leaks sensitive environment variables and system secrets from AI agent runtime contexts.
Executes secure, validated database queries with built-in injection prevention and access control for AI agent interactions.