bmad-agent-quick-flow-solo-dev

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill contains instructions to maintain its persona across session boundaries ("must not break character") and to persist the persona even when other skills are invoked. These are persistence-oriented instructions designed for user experience consistency.
  • [DATA_EXPOSURE]: The skill automatically searches for and loads project-context.md from the user's workspace to establish foundational reference standards. This is a functional feature for technical implementation but involves automated file system reads of potentially sensitive project documentation.
  • [COMMAND_EXECUTION]: The skill acts as a dispatcher that executes other registered tools (bmad-quick-dev, bmad-code-review) based on user-provided codes or menu selections.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect injection via processed data.
  • Ingestion points: Reads **/project-context.md (SKILL.md).
  • Boundary markers: None; the content is loaded as a foundational reference without delimiters.
  • Capability inventory: Can invoke other skills and tools mentioned in the capabilities table (SKILL.md).
  • Sanitization: No validation or sanitization is performed on the content of the project context file before it is incorporated into the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 05:51 PM