bmad-agent-dev
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes Python scripts such as resolve_customization.py and resolve_config.py from the project's _bmad/scripts/ directory using the uv tool to initialize its environment.
- [DYNAMIC_EXECUTION]: The agent processes and runs activation_steps_prepend and activation_steps_append defined in both its internal customize.toml and project-level TOML overrides to configure its behavior dynamically at startup.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from several project-local sources. 1. Ingestion points: customize.toml, team/user TOML overrides, file globs in persistent_facts, and development tickets like initiative-.md or story-.md. 2. Boundary markers: The skill does not define specific markers or guardrails for the ingested content. 3. Capability inventory: The skill can execute shell commands and activation steps based on these configurations. 4. Sanitization: The skill does not apply explicit sanitization to the ingested project files.
- [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of additional modules from the bmad-code-org GitHub organization using the npx skills add command, which are recognized vendor resources.
Audit Metadata