The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill demonstrates a vulnerability to indirect prompt injection where malicious instructions embedded in an input PRD could influence the generated UX specification and subsequent development prompts.
Ingestion points: The skill ingests untrusted Product Requirement Documents (PRDs) from users or other skills as described in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the ingested PRD content.
Capability inventory: The skill instructs the agent to use the Write tool to save the generated specification to the file system at design/06-ux-spec.md.
Sanitization: The instructions lack any sanitization, validation, or escaping logic for the PRD content before it is processed and written to the output file.

ux-specification