bmad-ux

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local Bash and Python scripts (wcag-checklist.sh, contrast-check.py, responsive-breakpoints.sh) to generate reference material and verify color contrast ratios. These scripts are statically provided within the skill's scripts directory and perform only mathematical calculations or text output.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8) due to the ingestion of external data for planning purposes:
  • Ingestion points: Reads project requirements and architecture from bmad-output/prd.md and bmad-output/architecture.md in SKILL.md.
  • Boundary markers: Absent; the skill does not instruct the agent to use delimiters or ignore embedded instructions when passing project data to subagents.
  • Capability inventory: Subprocess execution (Bash/Python) and file system access (Read, Write, Edit, Glob, Grep) within the project directory.
  • Sanitization: None; ingested data is directly interpolated into subagent prompt templates.
  • [SAFE]: No obfuscation, persistence mechanisms, privilege escalation, or unauthorized network operations were identified. The skill's stated purpose of UX planning aligns with its technical implementation and limited capability set.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 12:30 PM
Security Audit — agent-trust-hub — bmad-ux