tooluniverse-clinical-trial-matching
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources such as ClinicalTrials.gov and PubMed abstracts, which could be used for indirect prompt injection.
- Ingestion points: Data enters the skill context through tools like search_clinical_trials, get_clinical_trial_eligibility_criteria, and PubMed_search_articles.
- Boundary markers: There are no explicit instructions to the agent to treat external content as untrusted or to use specific delimiters to separate data from instructions.
- Capability inventory: The skill has the capability to write reports to the local file system and perform multiple network-based API calls to specialized bioinformatics tools.
- Sanitization: While the skill uses regular expressions to parse specific biomarkers, it does not sanitize or filter the broader descriptive text for potential instructions.
Audit Metadata