The Agent Skills Directory

[SAFE]: No security issues detected. The skill performs proofreading tasks on local manuscripts with a focus on American English conventions. It implements a non-negotiable human-in-the-loop process where every proposed edit is shown with context and must be explicitly approved by the user before the Edit tool is used. The skill is constrained to prose and specifically avoids modifying code blocks, LaTeX commands, or math mode, which reduces the risk of document corruption. No indicators of prompt injection, data exfiltration, or unauthorized command execution were found.

readability