The Agent Skills Directory

[SAFE]: No malicious instructions or patterns were identified. The skill is designed to improve code quality through standard refactoring techniques.\n- [COMMAND_EXECUTION]: The skill instructions utilize the Bash tool to run a local analysis script located at .claude/scripts/code_graph. This is a legitimate project-internal tool used to trace code dependencies and does not involve remote code downloads or privilege escalation.\n- [PROMPT_INJECTION]: The skill uses directive and reinforced language (e.g., 'MANDATORY', 'MUST ATTENTION', 'NON-NEGOTIABLE') to enforce quality standards and ensure the agent performs necessary verification steps. These instructions are focused on task performance and do not attempt to bypass core AI safety guidelines.\n- [DATA_EXFILTRATION]: There are no instructions or tool configurations that allow for network communication or the extraction of sensitive data. All file operations are scoped to the project's codebase for the intended purpose of refactoring code.

refactoring