tooluniverse-admet-prediction
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied chemical identifiers that are used as inputs for various scientific tools and APIs. \n- Ingestion points: Drug names and SMILES strings provided by the user in SKILL.md. \n- Boundary markers: The instructions do not specify the use of delimiters or boundary markers to isolate untrusted user data. \n- Capability inventory: The skill calls external bioinformatics APIs (PubChem, SwissADME, ChEMBL) and executes Python scripts via the shell. \n- Sanitization: No explicit sanitization or validation steps are defined for the incoming chemical strings. \n- [DYNAMIC_EXECUTION]: The skill instructs the agent to generate and run Python code using libraries like pandas and scipy to calculate ADMET scores and generate reports. This is a functional requirement for the skill but involves runtime code generation. \n- [COMMAND_EXECUTION]: The documentation provides a shell command to install the 'tooluniverse' package group, which is a standard procedure for setting up the skill's environment.
Audit Metadata