The Agent Skills Directory

[DYNAMIC_EXECUTION]: The assets/compliance_rules.json file contains functional Python code snippets within the check objects for various compliance standards (e.g., SOC2-Security-1, PCI-DSS-Network-1). These snippets are designed to be evaluated at runtime to perform specific compliance validations, representing an embedded code execution pattern.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external infrastructure configurations, creating an attack surface for indirect prompt injection.
Ingestion points: The skill processes local project files including Terraform configurations, Kubernetes manifests, and cloud-native IAM policies (referenced in SKILL.md).
Boundary markers: No specific delimiters or "ignore instructions" markers are defined for the agent when processing the content of these external files.
Capability inventory: The skill utilizes broad bash execution capabilities (Bash(cmd:*)) to run external scanners and has the ability to write remediation patches back to the filesystem.
Sanitization: There is no evidence of input validation or sanitization for the data ingested from the external infrastructure files before it is processed by the agent or used in command-line arguments.

checking-infrastructure-compliance