lit-review

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

• Third-party content exposure detected (high risk: 1.00). The skill's Step 2 "Systematic Search" explicitly requires using web search to find and verify papers (i.e., fetching and ingesting content from the open/public web and external websites) which the agent is expected to read and interpret as part of its workflow, exposing it to untrusted third-party content.