The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data by reading project documentation files and analyzing user-provided copy. This creates a surface where malicious instructions embedded in these files could attempt to influence the agent's behavior.
Ingestion points: The skill instructs the agent in SKILL.md to read local files such as PRODUCT.md, DESIGN.md, .impeccable.md, and .github/copilot-instructions.md to gather design context.
Boundary markers: There are no specific delimiters or instructions to ignore potential commands embedded in the processed text.
Capability inventory: The skill is restricted to providing text-based UX recommendations and does not possess capabilities for file modification, network communication, or shell command execution.
Sanitization: No validation or sanitization of the input text is performed before processing.

clarify