The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes findings from external AWS services to identify compliance issues. \n
Ingestion points: The get_findings API call is used in scripts/agent.py to retrieve findings from AWS Security Hub. \n
Boundary markers: The skill processes structured JSON findings but does not implement specific boundary markers for the findings' textual content. \n
Capability inventory: The skill can update finding statuses and modify S3 bucket configurations (blocking public access). \n
Sanitization: Employs simple keyword-based matching (e.g., "s3" and "public") on finding titles to trigger remediation actions.\n- [COMMAND_EXECUTION]: The skill documentation provides various AWS CLI commands for infrastructure management and security configuration. \n
Evidence: Multiple shell blocks in SKILL.md demonstrating aws securityhub and aws events usage. \n
Context: The commands are standard administrative operations required for the stated purpose of cloud security compliance.\n- [DYNAMIC_EXECUTION]: The skill documentation includes a step to generate a Python remediation script locally. \n
Evidence: A bash heredoc in SKILL.md creates a file at /tmp/remediate_s3.py. \n
Context: This is an instructional example for setting up automated remediation workflows and does not execute untrusted remote code.

implementing-aws-security-hub-compliance