review-deep
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
npx repomixto bundle the repository contents. This fetches therepomixpackage from the NPM registry at runtime.\n- [COMMAND_EXECUTION]: The skill executes various shell commands to manage temporary files and launch sub-processes, includinggit,rm,mkdir,npx, andopencode.\n- [DYNAMIC_EXECUTION]: Shell scripts for assembling results and orchestrator prompts are dynamically generated at runtime based on the target path and model labels.\n- [INDIRECT_PROMPT_INJECTION]:\n - Ingestion points: The skill processes the entire target codebase as input (SKILL.md, Step 2).\n
- Boundary markers: No explicit delimiters or instructions are provided to the subagents to ignore or isolate prompts contained within the source code being reviewed.\n
- Capability inventory: Subagents are granted access to the codebase and can write their findings to the filesystem.\n
- Sanitization: There is no evidence of filtering or sanitizing the code content before it is analyzed by the AI models.
Audit Metadata