domain-design

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

• Third-party content exposure detected (medium risk: 0.65). Yes—during the required “style guide gate” onboarding branch (SKILL.md:19-22), the workflow can fetch a user-provided website URL and extract palette/fonts from fetched page content (references/onboarding.md:41-45, 14-24), which is outsider-authored free text that may be ingested into the agent’s LLM context for analysis.