The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses uv run to execute local Python scripts (resolve_customization.py and resolve_config.py) located within the project's _bmad/scripts/ directory to manage its configuration and persona settings.
[DYNAMIC_EXECUTION]: The skill is designed to execute arbitrary instructions or commands provided in the activation_steps_prepend and activation_steps_append arrays within its configuration files. This allows for runtime execution of logic defined outside the core skill instructions. It also supports loading contents from files specified by glob patterns in the persistent_facts array.
[INDIRECT_PROMPT_INJECTION]: The agent's persona, principles, and activation steps are dynamically constructed by merging multiple local configuration files. If these files are modified by an untrusted party, they could influence the agent's behavior.
Ingestion points: customize.toml, {project-root}/_bmad/custom/{skill-name}.toml, and {project-root}/_bmad/custom/{skill-name}.user.toml.
Boundary markers: None identified in the instruction steps.
Capability inventory: Execution of scripts via uv run, execution of arbitrary strings in activation steps and menu prompts, and reading files via glob patterns.
Sanitization: No sanitization or validation of the merged configuration content is described before use in prompts or execution steps.
[EXTERNAL_DOWNLOADS]: The skill manifest identifies a GitHub repository under the vendor's organization as the source for updates.

bmad-agent-pm