next-theme-design

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions and test files (tests/test_design_package.py, tests/test_skill_package.py) involve the execution of the skill's own Python and Node.js scripts via the Bash tool and subprocess.run calls.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external websites, which serves as a potential injection vector.\n
  • Ingestion points: scripts/capture-page.js extracts text, link destinations, and accessible names from target URLs.\n
  • Boundary markers: The captured data is stored in structured JSON and Markdown files within a handoff package.\n
  • Capability inventory: The agent has access to Write, Edit, and Bash tools to manage the package files.\n
  • Sanitization: The Python scripts use standard JSON libraries for data serialization.\n- [EXTERNAL_DOWNLOADS]: The skill provides instructions in SKILL.md and README.md to install related tools from the NextCommerceCo organization on GitHub using npx and the ./skills.sh installer.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 05:52 AM