Explain any shell command offline: a plain-English, token-by-token breakdown of the program, its flags and operands — including pipes, redirects, subshells and common inline languages (sed/awk/jq) — plus a risk assessment. Use it to understand what a command line does before running or recommending it.

Safety-check a shell command BEFORE executing it. Returns a risk verdict (danger / caution / none) and plain-English warnings for destructive patterns: rm -rf /, curl | sudo bash, dd/mkfs/shred/wipefs to a disk device, chmod -R 777 /, git push --force, truncating /etc/passwd, fork bombs, kill -9 -1, find / -delete, and more. Ideal as a guard an AI agent calls before running shell commands.

Safety-scan a WHOLE shell script (multi-line text) BEFORE writing or running it. Runs the same offline danger engine as check_command_safety over every logical line (comments/shebangs stripped, backslash-continuations joined) and returns each destructive or risky command with its line number: rm -rf /, curl | sudo bash, dd/mkfs/shred to a device, chmod -R 777 /, git push --force, CI ${{ }}-injection sinks, and more. Ideal for an AI agent to pre-scan a script it just generated before saving or executing it.