improve-codebase-architecture
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local system commands including 'git log' to analyze project history and platform-specific commands ('open', 'xdg-open', 'start') to display the generated HTML report to the user.- [EXTERNAL_DOWNLOADS]: The generated architectural report template fetches the Tailwind CSS framework and Mermaid.js diagramming library from well-known content delivery networks (CDNs) at runtime.- [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by ingesting untrusted codebase data and rendering it into an HTML report with elevated visualization permissions.
- Ingestion points: Codebase file contents, commit history, and domain model definitions (SKILL.md).
- Boundary markers: No explicit sanitization or instructions to ignore embedded malicious strings during report generation were identified.
- Capability inventory: The skill can write files to the operating system's temporary directory and trigger shell commands to open those files in a web browser.
- Sanitization: The report configuration in 'HTML-REPORT.md' explicitly sets Mermaid's 'securityLevel' to 'loose', which disables certain security sandboxing and could allow malicious content within the analyzed codebase to execute scripts or render dangerous HTML in the user's browser context.
Audit Metadata