The Agent Skills Directory

[SAFE]: The skill implements a robust security guardrail through a mandatory credential boundary check defined in references/credential-boundary.md. It validates that the AWS principal used for the assessment lacks write permissions (e.g., Create, Update, Delete, Put) before executing any checks, effectively preventing accidental modifications to sensitive production environments.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute standard AWS CLI commands for infrastructure auditing. These commands are limited to read-only operations such as Describe, Get, and List, which are necessary to evaluate the compliance posture of services like S3, RDS, and IAM.
[EXTERNAL_DOWNLOADS]: The report generation script (scripts/generate-html-report.py) mentions an optional dependency on the markdown2 Python package. However, the script is designed with a minimal built-in fallback parser to avoid mandatory external dependencies and minimize the supply chain attack surface.
[DATA_EXFILTRATION]: No unauthorized network operations or data transmission patterns were identified. The skill processes assessment data locally and generates reports within the user's local directory structure.

ferpa-reviewer