Security audit
Security checks for vulnerabilities and agentic risk
This is a document-editing skill that rewrites AI-sounding text, with no evidence of hidden execution, network access, credential handling, or persistence.
Install only if you want an agent to rewrite user-provided or local document text to sound less AI-generated. Review proposed edits before saving or publishing, especially where tone, attribution, or factual precision matters.
The phrase "Or invoke directly when editing documents" does not specify exact trigger phrases, scope, or exclusion conditions for when the skill should activate. Because ordinary document editing is a very common context, this ambiguity could cause unintended invocations.
The description says to use the skill when "editing or reviewing text," which is a very broad context that overlaps with many ordinary writing tasks. It does not clearly distinguish when the skill should activate versus when a general editing skill should be used, nor does it provide exclusions or negative examples.
The manifest grants Write and Edit capabilities, and the instructions direct the skill to rewrite problematic sections and present a humanized version. However, the markdown does not explicitly warn users that the skill may substantially rewrite their text or recommend review before applying changes, which is a user-impacting behavior for markdown-scoped warning checks.
No suspicious patterns detected.
