improve-codebase-architecture
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes system-specific shell commands such as xdg-open, open, or start to launch the generated HTML report file stored in the temporary directory.
- [EXTERNAL_DOWNLOADS]: The generated report fetches external resources from well-known content delivery networks, including Tailwind CSS and jsDelivr (for Mermaid.js), to handle layout and rendering.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by reading and incorporating data from repository files like CONTEXT.md and ADRs. It lacks boundary markers to separate this untrusted data from its primary instructions and does not perform sanitization, allowing content in those files to potentially influence the agent's architectural conclusions.
- [COMMAND_EXECUTION]: The Mermaid.js configuration within the generated report uses securityLevel: 'loose', which can allow script execution within diagrams if the codebase metadata being rendered is maliciously crafted.
Audit Metadata