list-building

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

• Third-party content exposure detected (high risk: 1.00). The skill's required workflow (Steps 3-4 and "Browser Tools Used") instructs the agent to navigate to public LinkedIn search pages (linkedin.com/sales/search/people and linkedin.com/search/results/people/) and to use read_page to extract profile/search-result content, which are untrusted third-party, user-generated pages that the agent reads and acts on to build lists and drive follow-up actions.