brief-remote-friendliness

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md procedure (steps 3–5) explicitly instructs the agent to fetch and verbatim-quote content from public careers pages, engineering blogs, CEO/founder posts/podcasts, live job postings, and LinkedIn profiles and then use that content to decide candidate eligibility and update CRM records, which clearly ingests untrusted third-party content that could carry indirect prompt-injection.