Security audit
Security checks for vulnerabilities and agentic risk
This is a plain text-editing skill for making writing sound less AI-generated, with no hidden execution, network, credential, or persistence behavior found.
Install this only if you want the agent to rewrite prose to sound more natural and less AI-like. Review edits before accepting them, especially for factual, legal, academic, or brand-sensitive writing, because the skill may make tone stronger or more personal.
The phrase "Or invoke directly when editing documents" does not define specific trigger phrases, scope limits, or exclusion conditions. Because this README is a markdown file, vague activation language applies here and could lead to accidental or overly broad skill invocation.
The skill description is broad enough to match ordinary editing or reviewing requests, which can cause the agent to invoke this skill in contexts where the user did not specifically ask to humanize text. That increases the chance of unintended rewriting of user content, including changing tone, adding first-person phrasing, or altering meaning under the instruction to 'add soul' and 'have opinions.'
No suspicious patterns detected.
