brand-guidelines-builder

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs in "STEP 1 — GATHER BRAND INPUTS" that "If the user provides a URL, website, or existing doc — read it first and extract as much as possible before asking questions," meaning the agent will fetch and ingest arbitrary third-party web content provided by users and use it to drive brand decisions.