cli-reference

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides static documentation and usage examples for the claude command-line interface tool. It contains no executable code or malicious instructions.
  • [COMMAND_EXECUTION]: The documentation describes high-risk automation flags such as --dangerously-skip-permissions. This is identified as a functional requirement for non-interactive CI/CD environments and headless agentic workflows within the documented tool's ecosystem.
  • [PROMPT_INJECTION]: The guide details standard methods for customizing the model's system prompt using flags like --system-prompt and --append-system-prompt. These are legitimate configuration features of the tool and are not presented in a context that attempts to override the primary agent's safety protocols.
  • [EXTERNAL_DOWNLOADS]: Reference is made to the claude update command, which is the official mechanism for the tool to maintain itself.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 09:03 AM
Security Audit — agent-trust-hub — cli-reference