The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script located at {project-root}/_bmad/scripts/resolve_customization.py during activation and completion to handle configuration merging. This is part of the standard BMad framework workflow.
[DYNAMIC_EXECUTION]: The workflow instructions direct the agent to execute shell commands defined in the activation_steps_prepend and activation_steps_append arrays within the customize.toml and related override files. This allows for project-specific customization of the activation process.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external game design documents (GDD) and game briefs ({output_folder}/*gdd*.md, {output_folder}/*brief*.md) to inform the playtest plan, creating an attack surface for embedded instructions.
Ingestion points: SKILL.md (Workflow Architecture: Input Files).
Boundary markers: None specified for ingested documents.
Capability inventory: File system write (playtest-plan.md), subprocess execution (python3 script calls).
Sanitization: No explicit sanitization or filtering of ingested document content is described.

gds-playtest-plan