The Agent Skills Directory

[COMMAND_EXECUTION]: The Python script scripts/agent.py executes system commands using subprocess.run to interact with the aws and gcloud CLI utilities. This execution is performed to collect resource configuration and status information for security assessment purposes, which is consistent with the skill's primary function.
[INDIRECT_PROMPT_INJECTION]: The skill functions by ingesting and processing data from external cloud environments, which introduces a theoretical surface for indirect prompt injection if those environments contain malicious data.
Ingestion points: The scripts/agent.py script captures stdout from cloud discovery commands such as aws ec2 describe-verified-access-instances.
Boundary markers: No specific delimiters or instructions for the agent to ignore embedded commands are present in the script logic.
Capability inventory: The script has the capability to execute shell discovery commands via subprocess.run and write assessment reports to the local file system.
Sanitization: The script mitigates common command injection risks by passing arguments as a list to subprocess.run and utilizes json.loads to safely parse structured output from the CLI tools.

implementing-zero-trust-network-access