improve-codebase-architecture
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill generates HTML reports that reference external resources from well-known services, specifically fetching the Tailwind CSS framework and the Mermaid diagramming library from cdn.tailwindcss.com and cdn.jsdelivr.net. These are standard developer tools used for rendering visual components.
- [COMMAND_EXECUTION]: The skill executes shell commands to analyze the project's commit history (git log --oneline) and to open the generated report in the user's default browser using system-specific commands like xdg-open, open, or start.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the project's codebase (source files, commit history, CONTEXT.md, ADRs) as specified in the Explore section of SKILL.md. It lacks explicit boundary markers or sanitization to separate this data from its instructions. The skill has capabilities to write to the file system (modifying CONTEXT.md and writing to the /tmp directory) and execute system commands to display reports, creating an ingestion surface where malicious content in the scanned repository could influence agent behavior or report content.
- [DYNAMIC_EXECUTION]: The skill dynamically generates an HTML file in the system's temporary directory. The template in HTML-REPORT.md initializes the Mermaid library with securityLevel: 'loose', which enables advanced diagram features but increases the potential attack surface if the codebase content processed by the skill contains malicious scripts intended for execution in the user's browser environment.
Audit Metadata