The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script {project-root}/_bmad/scripts/resolve_customization.py using uv run to resolve its configuration during the activation phase. This script is part of the vendor-associated project structure.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from multiple external sources, which constitutes a surface for indirect prompt injection.
Ingestion points: The skill loads data from {project-root}/_bmad/cis/config.yaml, various .toml files for persona overrides, and content from files matched by glob patterns in {agent.persistent_facts}.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between its core instructions and the data loaded from these files.
Capability inventory: The skill possesses the ability to execute local scripts and invoke other skills in the system.
Sanitization: The instructions do not specify any validation or sanitization for the content ingested from the project files.

bmad-cis-agent-brainstorming-coach