The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to read and process the contents of a DESIGN.md file to determine typography and color schemes. This ingestion of untrusted external project data into the prompt context creates a potential attack surface for indirect prompt injection, where malicious instructions could be embedded in the design document to influence the agent's behavior. * Ingestion points: The file 'DESIGN.md' is consumed as input in the first step of the workflow described in SKILL.md. * Boundary markers: The instructions do not define clear delimiters or specific warnings to the agent to treat the design data as untrusted or to ignore embedded instructions. * Capability inventory: The skill produces web artifacts (HTML/CSS). * Sanitization: No input validation or filtering of the content within the design file is mentioned in the workflow.

motion-frames