The Agent Skills Directory

[SAFE]: The skill incorporates comprehensive security documentation, including a warning against combining dynamic shell command preprocessing with user arguments to prevent potential command injection.
[SAFE]: The defined workflows use legitimate tools such as git and grep for codebase analysis and duplicate detection, which are consistent with the skill's purpose of skill management.
[SAFE]: No evidence of obfuscation, hardcoded credentials, malicious persistence, or remote code execution was found in any of the skill's files or reference materials.
[SAFE]: While the skill ingests external input via arguments and existing skill files, the operations performed on this data are limited to safe reading and analysis, with the skill itself providing defensive documentation for authoring more secure agents.

writing-skills