The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs high-risk modifications to AWS infrastructure settings.\n
scripts/agent.py: The experiment_open_security_group function grants unrestricted SSH access (0.0.0.0/0) to security groups via ec2.authorize_security_group_ingress.\n
scripts/agent.py: The experiment_create_admin_user function creates a new IAM user and attaches the AdministratorAccess managed policy.\n
scripts/agent.py: The experiment_stop_cloudtrail function disables logging for AWS CloudTrail trails, potentially blinding security monitoring.\n- [EXTERNAL_DOWNLOADS]: The skill utilizes the boto3 library from Amazon Web Services to interact with cloud services.\n
scripts/agent.py: Imports and uses the boto3 Python package to execute API calls for resource manipulation and monitoring verification.\n- [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection via unsanitized data ingestion (Category 8).\n
Ingestion points: Command-line arguments --sg-id, --trail-name, and --profile in scripts/agent.py are used to target specific AWS resources.\n
Boundary markers: No boundary markers or instructions to ignore embedded commands are present in the processing logic.\n
Capability inventory: Includes the ability to modify security groups, IAM users, policies, and CloudTrail logging status across an AWS account.\n
Sanitization: No input validation or sanitization is performed on the resource identifiers before they are passed to AWS API calls.

implementing-security-chaos-engineering