The Agent Skills Directory

[SAFE]: The skill's primary script, scripts/prepare_ligand.py, uses standard scientific libraries such as RDKit and Meeko to process chemical data. It does not perform any network operations, access sensitive file paths, or attempt to execute arbitrary shell commands.
[SAFE]: Filename sanitization is implemented in the safe_stem function, which reduces risks associated with processing user-provided molecule names by removing special characters and spaces.
[SAFE]: No obfuscation, prompt injection attempts, or persistence mechanisms were detected in the instructions or scripts. The code is transparent and the metadata matches the functional behavior.
[SAFE]: The skill relies on well-known open-source chemistry packages including rdkit, meeko, and dimorphite_dl. Although the script contains a minor code defect (undefined variables _params_path and _config at the end of the main function), this does not present a security risk.

drug-ligand-prep