advpl-refactoring
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines clear refactoring patterns such as function extraction, simplifying conditionals, and removing dead code to improve codebase maintainability.
- [SAFE]: Includes a robust safety protocol requiring manual user approval before any code modifications are performed, ensuring human-in-the-loop verification.
- [SAFE]: Explicitly warns against modifying business logic or renaming variables with global scope (Public/Private) to avoid breaking external dependencies.
- [INDIRECT_PROMPT_INJECTION]: The skill inherently processes untrusted code files which represents a potential surface for indirect prompt injection via comments or metadata within the code.
- Ingestion points: The agent reads and analyzes ADVPL/TLPP source code files provided by the user (SKILL.md).
- Boundary markers: No explicit markers or 'ignore instructions' delimiters are defined for the code being processed.
- Capability inventory: The skill utilizes filesystem search (Grep), file reading, and file writing operations (SKILL.md).
- Sanitization: No specific sanitization or validation is applied to the input code prior to processing.
Audit Metadata