tooluniverse-gwas-study-explorer

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

• Third-party content exposure detected (low risk: 0.10). The runtime path in python_implementation.py ingests free text coming from search results—compare_gwas_studies() calls tu.run({"name":"gwas_search_studies", "arguments":{"disease_trait": trait, "size": max_studies}}) and then parses provider-returned text fields like initial_sample_size via _parse_sample_size()—but the only attacker-controlled “outsider text” is the user-supplied trait query, not arbitrary outsider-authored content submitted to the provider feeds/queues themselves.