The Agent Skills Directory

[DATA_EXFILTRATION] (SAFE): No sensitive data access or exfiltration patterns detected. The Terraform and Pulumi examples use best-practice variable interpolation and placeholders for bucket names and regions.
[REMOTE_CODE_EXECUTION] (SAFE): The skill does not download or execute code from remote sources. All logic is contained within the local validation script.
[COMMAND_EXECUTION] (SAFE): The scripts/validate.py script performs only passive file system checks (existence and directory listing). It does not use subprocess, os.system, or other functions to execute system commands.
[DYNAMIC_EXECUTION] (SAFE): The validation script uses yaml.safe_load() for parsing configuration files, which correctly prevents unsafe deserialization of arbitrary objects.
[PROMPT_INJECTION] (SAFE): No instructions attempting to override agent behavior, bypass safety filters, or extract system prompts were found in the markdown or metadata.
[CREDENTIALS_UNSAFE] (SAFE): No hardcoded API keys, tokens, or private secrets are present. Infrastructure examples correctly demonstrate the use of IAM roles and variables.

iac-automation