The Agent Skills Directory

[SAFE] (SAFE): The skill implements legitimate chemical informatics functionality using the industry-standard RDKit library. No patterns of prompt injection, data exfiltration, or malicious obfuscation were found.
[INDIRECT_PROMPT_INJECTION] (INFO): The skill processes external molecular data files (SDF, SMILES) via the process_file and load_molecules functions. While these are data ingestion points for untrusted content, the skill lacks high-risk capabilities such as network communication, arbitrary shell execution, or privilege escalation, which mitigates the risk of indirect prompt injection.
[COMMAND_EXECUTION] (SAFE): Command-line arguments are parsed using argparse and used for file path resolution and parameter setting. There is no evidence of unsafe string interpolation into shell commands or eval()/exec() calls.

rdkit