The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the xtb command-line utility and internal test scripts via subprocess.run. The commands are constructed using argument lists rather than raw shell strings, which is a secure implementation that mitigates command injection risks.\n- [SAFE]: The skill handles molecular data formats (SMILES, XYZ) using the RDKit library and dedicated parsers. There is no evidence of untrusted data being interpolated into agent prompts or used in a way that could lead to indirect prompt injection.\n- [SAFE]: No hardcoded credentials, sensitive file access (e.g., to .ssh or .env files), or unauthorized network exfiltration attempts were identified in the codebase.\n- [SAFE]: The skill's YAML frontmatter and documentation accurately describe its functionality, and no deceptive metadata or hidden instructions were found.

geometry-optimizer