The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Python subprocess module to execute external CLI tools like checkov and opa. Specifically, in the ci_compliance.py script, a directory path variable is passed as a command argument, which presents a risk of command injection if the agent handles unsanitized user or system input.
[DATA_EXFILTRATION]: The skill is designed to collect and transmit system metadata. Scripts such as collect_evidence.py and compliance_tests.py perform extensive read operations on AWS IAM and Kubernetes configurations. This data, including resource names and vulnerability findings, is transmitted to an external API endpoint defined by the COMPLIANCE_API_URL environment variable.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection.
Ingestion points: Untrusted data enters the context through system logs, application logs, and code review records as defined in references/audit-evidence.md and SKILL.md.
Boundary markers: Absent; there are no clear delimiters or instructions to ignore embedded commands in the processed data.
Capability inventory: The skill possesses powerful capabilities including shell command execution (subprocess.run), network operations (requests.post), and cloud resource access (boto3).
Sanitization: Absent; no validation or escaping of external log content is performed before processing.

enterprise-compliance-audit