The Agent Skills Directory

[SAFE]: The skill provides structured documentation, methodologies, and templates for information security management system auditing. The content is professionally focused on compliance and control testing without any signs of malicious intent.
[COMMAND_EXECUTION]: The skill includes a Python script scripts/isms_audit_scheduler.py for audit planning. This script uses standard libraries (argparse, csv, json) to process data locally. It does not use os.system, subprocess, or eval, making it safe for execution in the intended context.
[INDIRECT_PROMPT_INJECTION]: The audit scheduler script processes external data (CSV files), which is a common pattern for data processing skills. While this represents a theoretical surface for indirect prompt injection if malicious data is provided in the input CSV, the risk is negligible given the script's limited capabilities.
Ingestion points: scripts/isms_audit_scheduler.py reads user-supplied CSV files via the --controls parameter.
Boundary markers: No specific delimiters are applied to the input data within the script's output.
Capability inventory: The script is restricted to local file I/O (reading CSV, writing JSON/Markdown) and does not possess network or advanced system capabilities.
Sanitization: The script performs standard CSV parsing and string formatting without specific safety filtering for instructions embedded in data.

isms-audit-expert