Text Humanizer Instruction Based

Security checks across static analysis, malware telemetry, and agentic risk

Overview

This instruction-only skill has no code or credential access, but it is designed to make AI-written text appear human-authored, including in academic and job-application contexts.

This skill is operationally low-risk because it is instruction-only and requests no system access. Review it carefully if you work in academic, hiring, publishing, or policy-governed settings, because its purpose is to disguise AI-writing patterns and make text sound personally human-authored.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Using this skill could help present AI-assisted writing as personally authored human writing, which may violate school, employer, platform, or publication rules.

Why it was flagged

The skill explicitly targets removal of AI-authorship signals and includes contexts where readers may rely on authentic human authorship, such as academic work and job applications.

Skill content
rewrites them to sound like a human wrote them ... "make this sound like I wrote it" ... "job application, article, academic work, or social post"
Recommendation

Use only where rewriting for style is allowed, and do not use it to hide AI assistance where disclosure or original authorship is expected.

What this means

A normal editing request may receive a more opinionated de-AI rewrite than the user expected.

Why it was flagged

The invocation guidance can broaden ordinary editing requests into AI-tell removal without explicitly showing the user what was changed unless they ask for a report.

Skill content
Pastes text and asks to "rewrite," "clean up," or "edit" it and the text shows two or more patterns ... apply the fixes silently
Recommendation

Confirm whether the user wants AI-style removal, ordinary copyediting, or a change report before applying broad silent rewrites.