code-review
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE] (SAFE): The skill consists entirely of instructional Markdown content. There are no executable scripts (Python, JavaScript, Shell) or binaries associated with this skill.
- [EXTERNAL_DOWNLOADS] (SAFE): The source is identified as 'anthropics/claude-code', which belongs to a trusted organization. No remote dependencies or external scripts are fetched.
- [COMMAND_EXECUTION] (SAFE): No system commands, shell executions, or subprocess spawns are defined in the instructions.
- [DATA_EXFILTRATION] (SAFE): The skill does not contain any code to access sensitive files or perform network operations.
- [PROMPT_INJECTION] (SAFE): The provided text is purely instructional for the task of code review and does not contain any patterns designed to override agent safety or system instructions.
Audit Metadata