aura-asset-images

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow described in SKILL.md and demo/PROMPT.md instructs the agent to open Aura’s searchable asset pages (https://www.aura.build/assets), where an outsider (Aura users) could potentially submit poison text via tags/search results that the agent ingests before selecting specific items.