The Agent Skills Directory

[PROMPT_INJECTION]: The skill operates by ingesting content from various project directories, which constitutes an indirect prompt injection surface if those files contain malicious instructions.
Ingestion points: Content is read from src/, assets/, design/gdd/, and design/narrative/ across multiple modes (scan, brief, cultural-review, vo-pipeline, qa).
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when processing external text for report generation or translation extraction.
Capability inventory: The skill possesses the ability to write files via the Write tool, execute shell commands (specifically grep and glob patterns) via the Bash tool, and spawn new agent instances via the Task tool.
Sanitization: No explicit sanitization, escaping, or validation logic is defined for the content extracted from the project repository before it is processed by the agent or written to new files.

localize