cli-developer

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill inherently processes untrusted data through command-line arguments, environment variables, and configuration files.
  • Ingestion points: User input via arguments, flags, interactive prompts (inquirer, questionary), and configuration files (.json, .yml).
  • Boundary markers: Not explicitly defined for LLM context, but the skill follows standard CLI input parsing protocols.
  • Capability inventory: File system access (fs-extra, os, path), subprocess execution (execa, cobra.Command.RunE), and terminal UI manipulation.
  • Sanitization: The skill contains explicit 'MUST DO' instructions to 'Validate user input early' in the constraints section of SKILL.md.
  • [COMMAND_EXECUTION]: Provides guidance on using execa (Node.js) and cobra (Go) for executing commands. These are standard tools for CLI development and are documented with appropriate error handling and signal management.
  • [DATA_EXPOSURE]: Mentions the use of credentials.json for sensitive data but correctly identifies the need for restrictive file permissions (mode 600) as a security best practice.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 02:22 AM