The Agent Skills Directory

[SAFE]: No malicious patterns, such as prompt injection, data exfiltration, or persistence mechanisms, were detected in the skill's instructions or scripts.
[SAFE]: The skill proactively recommends security best practices, specifically advising users to avoid loading Python pickle files from untrusted sources and suggesting safer data interchange formats like SMILES and SDF.
[COMMAND_EXECUTION]: Shell commands are restricted to standard environment setup via uv or conda and the execution of the included Python utility scripts, which is appropriate for a cheminformatics toolkit.
[EXTERNAL_DOWNLOADS]: External dependencies are sourced from well-known and reputable package registries, including PyPI and the conda-forge channel.

rdkit