The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script using uv run. These scripts are located within the project's own _bmad directory and are used for resolving workflow customizations.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external Markdown files (epics) and TOML configuration files. It lacks explicit boundary markers or sanitization for these inputs, which could allow maliciously crafted content in epic files to influence agent behavior.
[DYNAMIC_EXECUTION]: The workflow allows for the execution of dynamic steps defined in configuration files (activation_steps_prepend, activation_steps_append, and on_complete). This provides a mechanism for runtime behavior modification based on the contents of the local file system.

gds-sprint-planning