The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill parses and analyzes content from markdown files (Game Design Documents and Architectural Decision Records) to determine the status of architectural decisions. Maliciously formatted content within these files could potentially influence the agent's analysis or reporting outcome.
Ingestion points: Reads user-provided GDD files and all ADR files in the docs/architecture/ directory.
Boundary markers: The skill does not implement specific delimiters or safety instructions to isolate the data being read from the operational instructions.
Capability inventory: The skill utilizes Bash (for git operations), Write (for document updates), and Task (for agent orchestration).
Sanitization: There is no evidence of sanitization or filtering of the text extracted from documentation before it is processed by the model.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute git show commands. This is used legitimately to retrieve previous versions of documents for comparison but represents a direct shell interaction path.

propagate-design-change