ontology-term-resolution

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow performs live HTTP requests to the outsider-authored public EBI OLS4 API (e.g., scripts/resolve_terms.py → ols_client.search()/term_detail() → _request() which urlopens JSON and loads it), and the returned JSON contains readable free-text labels/synonyms that are then used/serialized into the LLM context/results.