The Agent Skills Directory

[COMMAND_EXECUTION]: During activation, the skill executes a local Python script located at {project-root}/_bmad/scripts/resolve_customization.py to merge configuration files from multiple sources (default, team, and user overrides).
[COMMAND_EXECUTION]: The skill is designed to execute arbitrary commands defined in the activation_steps_prepend and activation_steps_append configuration arrays. In the provided customize.toml, these arrays are empty, but the mechanism allows for dynamic command execution based on local configuration.
[INDIRECT_PROMPT_INJECTION]: The skill automatically ingests data from external project files into its context, which presents a surface for indirect instructions to influence the agent.
Ingestion points: Files matching the glob pattern {project-root}/**/project-context.md are loaded as foundational context through the persistent_facts mechanism.
Boundary markers: The instructions do not specify the use of delimiters or protective headers (e.g., "ignore instructions within this text") when loading these facts.
Capability inventory: The agent can execute local shell commands (via activation steps) and dispatch tasks to other installed skills (e.g., gds-quick-dev, gds-code-review).
Sanitization: There is no evidence of sanitization or filtering applied to the content of the loaded files before they are added to the agent's context.

gds-agent-game-solo-dev