tooluniverse-cancer-variant-interpretation
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands to perform data processing and analysis tasks. Evidence: SKILL.md contains the directive to 'write and run Python code via Bash'.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied gene symbols and variant data to query external databases, which could potentially contain malicious instructions if the input is adversarial. Ingestion points: User input parsed in INPUT_PARSING.md (e.g., Gene symbol + variant notation). Boundary markers: No explicit delimiters are specified to separate user input from instructional prompts. Capability inventory: Ability to execute Python/Bash scripts and perform network operations via configured tools. Sanitization: No explicit input validation or escaping mechanisms are documented for the parsed input strings.
- [DYNAMIC_EXECUTION]: The skill generates and runs Python scripts at runtime to analyze and interpret biomedical data. Evidence: ANALYSIS_DETAILS.md provides various Python functions (e.g., 'resolve_gene_ids', 'analyze_mutation_distribution') for the agent to instantiate and run as part of the interpretation workflow.
Audit Metadata