The Agent Skills Directory

[SAFE]: The skill is entirely documentation-based, consisting of markdown files that guide AI agents and developers in writing safe shell scripts.
[SAFE]: Automated alerts regarding destructive commands (e.g., rm -rf /) in references/err-strict-mode.md and references/safety-command-injection.md are benign. These commands are part of educational 'Incorrect' examples used to illustrate security vulnerabilities like the 'unset variable' bug and command injection.
[SAFE]: No evidence of prompt injection, data exfiltration, or obfuscation was found in any of the 55 analyzed files.
[SAFE]: The skill references only established and trusted external sources, such as the Google Shell Style Guide, ShellCheck, and official GNU documentation.
[SAFE]: There are no suspicious network operations or dynamic context injection patterns (the !command syntax) used in the skill configuration.

shell