The Agent Skills Directory

[PROMPT_INJECTION]: The skill employs a persistence mechanism using a local system file (.interface-design/system.md) to store and retrieve design patterns and tokens. By saving user-influenced design choices and automatically applying them in future sessions, the skill creates an attack surface for indirect prompt injection. A malicious user could provide design constraints that are saved as patterns and later interpreted as behavioral instructions, potentially leading to a persistent compromise of the agent's logic. * Ingestion points: User-provided design requirements and interactive feedback stored in session context. * Boundary markers: No specific boundary markers or ignore instructions directives are present in the logic for reading or writing the system file. * Capability inventory: The skill utilizes file read and write operations on the .interface-design/system.md file. * Sanitization: The skill lacks defined validation or sanitization protocols for the content extracted from user prompts and saved to the system configuration.

interface-design