The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (merge-config.py, merge-help-csv.py, cleanup-legacy.py) using uv run and shell commands (mkdir -p) to configure the project environment. These scripts are used for merging configuration files and performing directory cleanup. Path validation is implemented within the scripts to ensure that filesystem arguments are resolved correctly before execution.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a data processor, ingesting module metadata and user configuration to generate persistent project files. This represents a standard injection surface for setup-oriented skills. Ingestion points: Reads metadata from assets/module.yaml and assets/module-help.csv, collects user-supplied configuration values, and parses legacy configuration files from the project directory. Boundary markers: Instructions provide logic for mapping values and establishing default priorities, though no explicit boundary delimiters are used in the generated files. Capability inventory: The skill has the capability to write configuration files to the _bmad/ directory, create output directories, and delete specified legacy subdirectories (shutil.rmtree). Sanitization: The scripts utilize yaml.safe_load for parsing and standard libraries for structured data output, reducing risks associated with malformed data.

bmad-bmb-setup