bmad-agent-pm

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Python scripts using uv run to resolve project-specific customizations, configurations, and to list available templates. These scripts are located within the user's project structure under the _bmad/ directory.
  • uv run {project-root}/_bmad/scripts/resolve_customization.py
  • uv run {project-root}/_bmad/scripts/resolve_config.py
  • uv run {project-root}/_bmad/scripts/list_templates.py
  • uv run {project-root}/_bmad/method/scripts/tickets.py
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for the agent to suggest installing the core bmad skill from the vendor's official GitHub repository if the environment setup is incomplete.
  • npx skills add bmad-code-org/BMAD-METHOD --skill bmad
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external project files and user-provided inputs to generate structured planning documents, representing an indirect injection attack surface.
  • Ingestion points: The skill reads persistent_facts from files or globs within the project root and parses content from initiative or epic tickets.
  • Boundary markers: No explicit boundary markers or delimiters for untrusted content are defined in the instructions; the skill relies on the downstream bmad-facilitate skill for content processing.
  • Capability inventory: The skill has capabilities to execute local scripts and invoke external skills.
  • Sanitization: No explicit sanitization or validation of the processed data is performed within this skill's instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 10:09 PM