improve-codebase-architecture

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses git log to identify codebase 'hot spots' by analyzing recent commit history.- [COMMAND_EXECUTION]: It executes platform-specific commands such as xdg-open, open, or start to display the generated HTML report to the user.- [EXTERNAL_DOWNLOADS]: The generated HTML report fetches styling and diagramming libraries from well-known and trusted services, specifically Tailwind CSS via tailwindcss.com and Mermaid via jsdelivr.net.- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted codebase data (file paths, module names) and includes them in a generated HTML report for visualization.\n
  • Ingestion points: Local codebase directory and Git commit logs.\n
  • Boundary markers: Not explicitly defined for HTML output generation.\n
  • Capability inventory: The agent can write to the OS temporary directory and execute local commands to open files.\n
  • Sanitization: The instructions do not specify sanitization for codebase strings before they are rendered in the Mermaid diagrams, though the risk is localized to the user's browser.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 01:30 AM
Security Audit — agent-trust-hub — improve-codebase-architecture