The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the aws CLI to retrieve resource configurations and utilization metrics from CloudWatch. These operations are restricted to read-only metadata and metrics access.
[DYNAMIC_EXECUTION]: Utilization reports are generated by piping AWS CLI output into python3 -c for data processing. This is a local execution pattern used to handle JSON data and perform calculations (e.g., savings estimates and threshold checks).
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from the AWS environment, such as Instance IDs and Resource Tags. Security risks are mitigated by using json.load for parsing and subprocess.run with argument lists (rather than shell strings) for subsequent CLI calls, preventing data-driven command injection.

aws-rightsizing