artifacts-builder

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): The skill presents an Indirect Prompt Injection surface (Category 8) by design, as it generates code artifacts from user descriptions. Ingestion points: User-provided interface descriptions and layout requests within the chat context. Boundary markers: Absent in the SKILL.md definition. Capability inventory: Generation of executable frontend React, Tailwind, and shadcn/ui code. Sanitization: No validation or sanitization logic is specified in SKILL.md to filter instructions within user data.
  • [COMMAND_EXECUTION] (LOW): The skill's verification metadata contains a reference to 'node .claude/scripts/sfc_lint.mjs'. This is a static analysis capability used for linting during development and does not represent an active runtime risk during user interactions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 08:24 PM