The Agent Skills Directory

[SAFE]: The skill provides a compliance expert persona designed to assist with regulatory audits and implementation guidance. All provided code snippets are pedagogical examples for implementing security controls such as encryption, access management, and audit logging.
[SAFE]: Code examples in resources/implementation-playbook.md use industry-standard libraries (e.g., Python's hashlib, json, uuid, datetime) and refer to security best practices like FIPS 140-2 and TLS 1.2+.
[SAFE]: The GitHub Action workflow provided as a template uses official actions (actions/checkout, actions/upload-artifact) and follows standard CI/CD patterns for security scanning.
[SAFE]: Placeholders such as SECRET_SALT and SECRET_KEY are used in logic examples to indicate where user-defined configuration should be placed, which is standard practice for template documentation.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied business information through $ARGUMENTS to generate compliance advice. While this is a common attack surface for indirect prompt injection, the skill includes instructions for the agent to clarify goals and validate outcomes, which serves as a basic manual safeguard. There are no automated tools or executable capabilities attached to this ingestion point that would elevate the risk.

security-compliance-compliance-check