bmad-check-implementation-readiness

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script (_bmad/scripts/resolve_customization.py) to resolve configuration and customization settings from TOML files during activation and upon completion. This is a standard part of the vendor's workflow management.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it reads and extracts content from untrusted project artifacts (PRDs, Architecture, Epics, UX docs) to generate assessment reports.\n
  • Ingestion points: Project documents found in the {planning_artifacts} directory (e.g., *prd*.md, *epic*.md, *ux*.md).\n
  • Boundary markers: No specific delimiters or safety warnings are used to isolate the content of analyzed documents from the agent's instructions.\n
  • Capability inventory: The agent can execute local shell commands (Python scripts), read project files, and write report artifacts.\n
  • Sanitization: There is no evidence of sanitization or filtering applied to the text extracted from external documents before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 05:26 PM