bmad-create-ux-design

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill incorporates an indirect prompt injection surface through its core functionality of reading and synthesizing external project documents.\n
  • Ingestion points: steps/step-01-init.md performs discovery and loading of markdown files (e.g., *brief*.md, *prd*.md) from project-specific directories such as {product_knowledge} and docs.\n
  • Boundary markers: The agent implements a control mechanism by requiring the user to explicitly confirm which discovered files should be loaded into the workflow context.\n
  • Capability inventory: The skill is capable of reading local project files (step-01), appending content to the specification document across multiple steps (step-02 through step-13), and generating HTML files for visual representations (step-08, step-09).\n
  • Sanitization: The skill does not use specific sanitization techniques or instruction-ignoring delimiters for the content read from external documents.\n- [COMMAND_EXECUTION]: The workflow performs various file operations, including copying a template to create a new UX specification and appending generated sections to that document. It also creates HTML files on the local filesystem to showcase color themes and design mockups.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 03:03 PM