Skills
skills/bmad-labs/skills/bmad-agent-architect/Gen Agent Trust Hub

bmad-agent-architect

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a Python script (resolve_customization.py) located in the project root to manage configuration merging during activation.
  • [COMMAND_EXECUTION]: The activation workflow involves executing steps defined in activation_steps_prepend and activation_steps_append from the customize.toml configuration and its overrides.
  • [COMMAND_EXECUTION]: The agent's menu supports a prompt action type that executes raw instruction strings defined in the configuration.
  • [PROMPT_INJECTION]: The skill loads arbitrary project context files (via {project-root}/**/project-context.md) into its persistent memory, creating an attack surface for indirect prompt injection where malicious content in these files could influence agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 05:25 PM