The Agent Skills Directory

[Remote Code Execution] (SAFE): No evidence of downloading or executing code from remote or untrusted sources. All operations occur locally.\n- [Data Exfiltration] (SAFE): The skill does not perform network requests or access sensitive user data like credentials or SSH keys.\n- [Command Execution] (SAFE): Command execution is limited to local Python scripts for geometric modeling. The provided workflow for custom designs uses safe modeling templates.\n- [Obfuscation] (SAFE): All files are written in clear, human-readable text without any encoding, zero-width characters, or homoglyphs.\n- [Prompt Injection] (SAFE): The instructions in the markdown files are strictly aligned with the stated purpose of CAD generation and do not contain safety bypass attempts.

stl-generator