bmad-brainstorming

Pass

Audited by Gen Agent Trust Hub on Mar 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands, specifically 'mkdir' and 'cp', during the session setup phase to create directories and initialize session documents based on internal configuration.
  • [PROMPT_INJECTION]: An indirect prompt injection surface is present due to the ingestion of an external 'context_file' to provide project-specific guidance. 1. Ingestion points: Workflow setup in 'step-01-session-setup.md' reads the external file. 2. Boundary markers: None identified; the context content is integrated directly into the session overview without delimiters. 3. Capability inventory: The skill can write to the local filesystem and invoke additional vendor-specific skills (e.g., bmad-advanced-elicitation). 4. Sanitization: No sanitization or filtering logic is present to prevent the agent from following instructions embedded within the untrusted context file.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 21, 2026, 11:47 AM