openalex

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and ingests results from the public OpenAlex API (via tools/openalex_fetch.py and the "Execute Search" step) and then parses and presents third-party fields like abstracts, OA URLs, topics and keywords (Step 4/Step 5), so untrusted public content can be read and influence follow-up actions/queries.