The Agent Skills Directory

[SAFE]: The skill provides comprehensive templates and reference materials for compensation planning. Analysis of all markdown files and the included validation script confirms the behavior is strictly limited to text generation and local validation of generated content.
[COMMAND_EXECUTION]: The scripts/check-output.py script is a local utility meant to be invoked by the agent to validate deliverables against a checklist. It uses standard library modules (re, sys, pathlib) to perform static text analysis and does not involve subprocess calls, shell execution, or dynamic code evaluation.
[EXTERNAL_DOWNLOADS]: The skill references various external benchmarks and articles from well-known platforms (e.g., LinkedIn, Apple Podcasts, SaaStr) and the author's own GitHub repositories. All external references are informational and do not involve remote code execution or automated downloads.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests user requirements to generate compensation plans, it lacks capabilities to execute shell commands, perform network operations, or write to sensitive system paths, effectively neutralizing the risk of indirect injection attacks.

executive-compensation