Admet Prediction
PassAudited by ClawScan on May 1, 2026.
Overview
The skill appears to be a coherent ADMET prediction helper with only purpose-aligned local file processing and package setup notes.
This looks safe to use for local ADMET screening. Before installing, verify the Python dependencies and run the script in a virtual environment; only provide compound files you are comfortable processing locally.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Installing the wrong or untrusted package could affect the local Python environment.
The skill asks users to install external Python packages. This is expected for cheminformatics prediction, but users should verify package names and sources before installing.
pip install rdkit # Optional for advanced models pip install deepchem admet-x
Install dependencies from trusted package indexes or project documentation, preferably in an isolated virtual environment.
The skill may process local compound library files that could contain proprietary research data.
The script reads user-provided local compound files for batch prediction. This is purpose-aligned and scoped to user-specified inputs.
def _read_compounds(self, input_file: str) -> List[tuple]: ... Chem.SDMolSupplier(input_file) ... with open(input_file) as f:
Only point the tool at files you intend to analyze, and avoid sharing generated outputs if they contain confidential compounds.