personal-tool-builder
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill instructs the agent on how to build tools securely, specifically highlighting risks like hardcoded credentials and offering remediations such as using environment variables and gitignored config files.\n- [COMMAND_EXECUTION]: Provides example code for checking system dependencies using
child_process.execSyncto execute shell commands likewhich. These examples are appropriately scoped for the purpose of validating local environments.\n- [SAFE]: Promotes binding internal web interfaces to localhost (127.0.0.1) to avoid accidental network exposure of personal data and services.\n- [SAFE]: All file system access patterns are intended for local storage within the user's home directory, following standard application data conventions.
Audit Metadata