tooluniverse-acmg-variant-classification

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources such as PubMed, ClinVar, and UniProt which could theoretically contain malicious instructions designed to influence the agent's reasoning.
  • Ingestion points: Data fetched from external tools including PubMed_search_articles, ClinVar_get_variant_details, and UniProt_get_function_by_accession (SKILL.md).
  • Boundary markers: The skill uses structured Markdown templates for the final report, providing some structural separation, but does not explicitly instruction the agent to ignore instructions embedded within the fetched data.
  • Capability inventory: The skill uses a variety of database lookup tools and a mechanism explanation tool; it does not contain capabilities for arbitrary shell command execution or persistent file system modifications.
  • Sanitization: The instructions do not specify sanitization or filtering logic for text fetched from external biological and clinical databases.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 06:31 AM
Security Audit — agent-trust-hub — tooluniverse-acmg-variant-classification