The Agent Skills Directory

[SAFE]: The skill is entirely instructional, providing a methodology for software debugging and patching. It contains no scripts, binaries, or executable code.
[SAFE]: The configuration explicitly disables autonomous model invocation (disable-model-invocation: true) and prevents implicit activation (allow_implicit_invocation: false), ensuring the skill only runs with direct user oversight.
[SAFE]: External URLs found in the documentation are for licensing attribution and link to official GitHub repositories, posing no security threat.

surgical-patch