The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from several local design files (game-concept.md, player-journey.md, interaction-patterns.md, and accessibility-requirements.md) to inform the instructions for spawned subagents.
Ingestion points: Multiple design and requirements markdown files are read during Phase 1a and Phase 3.
Boundary markers: The instructions do not specify using delimiters or 'ignore embedded instructions' warnings when passing content from these files to subagents.
Capability inventory: The skill uses the Task tool to spawn subagents and allows Bash execution, creating a potential path for instructions hidden in design files to influence command execution.
Sanitization: No explicit sanitization or validation of the content of these design files is performed before interpolation into subagent prompts.

team-ui