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 uses platform-specific commands such as xdg-open, open, or start to automatically open a generated HTML report in the user's default browser.
  • [EXTERNAL_DOWNLOADS]: The generated architectural report references external scripts from cdn.tailwindcss.com and cdn.jsdelivr.net (JSDelivr) for styling and rendering diagrams. These are well-known and established services.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted data from the codebase being analyzed (such as file names, code content, and git history) and incorporates it into the generated HTML report.
  • Ingestion points: Local source code files, directory structures, and output from git log.
  • Boundary markers: None specified for the HTML report generation phase.
  • Capability inventory: File system write access to the OS temporary directory and shell execution to open the generated report.
  • Sanitization: The skill explicitly configures the Mermaid diagramming tool with securityLevel: "loose". This configuration disables default security sanitization in the diagrams, which could allow raw HTML or JavaScript payloads found in the analyzed codebase (e.g., in a malicious file name or code comment) to execute in the user's browser context when the report is opened.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 10:28 AM
Security Audit — agent-trust-hub — improve-codebase-architecture