The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a Python script located at {project-root}/_bmad/scripts/resolve_customization.py during both the activation phase and upon completion. This script is a vendor-provided tool intended for environment resolution.
[COMMAND_EXECUTION]: The skill is designed to execute arbitrary instructions defined in configuration files (customize.toml and its overrides) via fields like activation_steps_prepend, activation_steps_append, and on_complete. These are executed as terminal instructions, providing a mechanism for dynamic command execution.
[PROMPT_INJECTION]: The skill implements an indirect prompt injection surface by aggregating data from external artifacts (Game Design Documents, Architecture docs, and Epics) and interpolating them into a master story file used by subsequent agents.
Ingestion points: Files including sprint-status.yaml, epics.md, gdd.md, architecture.md, and project-context.md.
Boundary markers: The output is structured using a template, but there are no specific delimitations or instructions to ignore malicious directives embedded within the ingested source documents.
Capability inventory: The skill has the ability to read and write files and execute shell commands/Python scripts.
Sanitization: No evidence of input validation, filtering, or sanitization of the artifact content is found before it is processed and stored.

gds-create-story