The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a Python script located in the project workspace (_bmad/scripts/resolve_customization.py) during its activation sequence to merge configuration settings.
[COMMAND_EXECUTION]: The instructions mandate the execution of dynamic steps defined in the activation_steps_prepend and activation_steps_append configuration arrays, as well as prompts defined in the agent.menu section.
[PROMPT_INJECTION]: The skill automatically ingests external data from the project root into the AI's core context, creating a surface for indirect prompt injection.
Ingestion points: Automatically loads all files matching the glob pattern {project-root}/**/project-context.md as well as the {project-root}/_bmad/bmm/config.yaml file.
Boundary markers: No specific delimiters or safety instructions are used to isolate the loaded file content from the agent's instructions.
Capability inventory: The agent has the ability to execute shell commands (via Python) and process dynamic menu prompts defined in local configuration.
Sanitization: No sanitization or validation of the loaded file content is performed; data is treated as foundational facts.

bmad-agent-ux-designer