claude-code
Pass
Audited by Gen Agent Trust Hub on Apr 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill references the installation of @anthropic-ai/claude-code and various Model Context Protocol (MCP) servers from official and well-known registries. These are documented as standard requirements for the tool.
- [COMMAND_EXECUTION]: Provides detailed guidance on orchestrating a terminal-based interface using tmux and automation flags like --dangerously-skip-permissions. These instructions are provided in the context of developer workflow automation.
- [SAFE]: A signature match for a fork bomb pattern was identified within a code block that demonstrates how to implement security hooks to block such malicious commands. This is a defensive configuration example.
- [SAFE]: Authentication instructions involving ANTHROPIC_API_KEY follow the official documentation for the CLI tool and do not involve credential exfiltration.
Audit Metadata