The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (fitness-config.py and overall.py) for configuration management and score aggregation. These scripts use standard libraries and do not perform network operations or access sensitive system files outside the project scope.
[REMOTE_CODE_EXECUTION]: The skill launches sub-agents using established internal task tools (/review:review-architecture, etc.). It does not download or execute code from untrusted external sources.
[DATA_EXPOSURE]: The instructions explicitly forbid hardcoding weights or secrets, directing the agent to refer to sensitive items by file and line number only.
[DYNAMIC_CONTEXT_INJECTION]: The skill uses the ${CLAUDE_SKILL_DIR} variable to resolve script paths locally. This is a standard platform feature for locating bundled resources and does not involve user-controlled command injection.

review-full