The Agent Skills Directory

[SAFE]: The skill consists of markdown-based instructions for architectural and business analysis. It does not contain any executable code, scripts, or system-level commands.
[DATA_EXFILTRATION]: All analysis and documentation activities are confined to the local project workspace. No network calls or data transmission to external domains are present.
[PROMPT_INJECTION]: While the skill is designed to process external code which could contain malicious injection strings, the risk is negligible because the skill lacks hazardous capabilities like code execution or network connectivity. Furthermore, the iterative cycle requires user confirmation before any documentation is persisted, ensuring human-in-the-loop oversight.

codebase-analysis