The Agent Skills Directory

[SAFE]: The skill provides a methodology and a Python tool (engagement_planner.py) for planning authorized security simulations. The tool calculates effort scores and organizes techniques into kill-chain phases without performing any actual offensive actions.
[COMMAND_EXECUTION]: The skill instructions and documentation provide example command-line usage of the included script. These commands are local, diagnostic, and do not involve unauthorized privilege escalation or dangerous shell operations.
[DATA_EXPOSURE_&_EXFILTRATION]: No hardcoded credentials, sensitive file path access, or network operations were found in the provided scripts or markdown files.
[INDIRECT_PROMPT_INJECTION]: While the engagement_planner.py script accepts user-supplied technique IDs and crown jewel labels, it lacks dangerous capabilities such as file system writes, network connectivity, or subprocess execution, rendering the surface for indirect injection safe.
[DYNAMIC_EXECUTION]: The Python script uses standard data structures and logic for planning. No use of eval(), exec(), or dynamic loading from untrusted sources was identified.

red-team