The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute local utility scripts (e.g., resolve_customization.py, resolve_config.py, memlog.py) using uv run. These commands are parameterized by environment-specific placeholders such as {project-root} and {skill-root}.
[DYNAMIC_EXECUTION]: The resolve_personas.py script dynamically constructs and executes command lists to invoke other Python scripts using the current Python interpreter (sys.executable). It targets specific paths within the project's internal _bmad directory.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied ideas and existing project files to provide feedback. It lacks explicit instruction-level delimiters or sanitization steps for handling potentially malicious content within these external inputs.
Ingestion points: User-provided subject ideas and goal descriptions; external project files and materials referenced during the session.
Boundary markers: None identified in the instructions for processing external project files.
Capability inventory: Local command execution via uv run and subprocess.run for logging, configuration, and persona resolution.
Sanitization: No specific sanitization or validation logic is applied to the content of ingested project files or user ideas before they are processed by the agent.

bmad-forge-idea