Humanize AI
PassAudited by ClawScan on May 10, 2026.
Overview
This skill is a local text-cleanup tool with no network or credential use, but users should review batch overwrites and the detector-bypass use case carefully.
This appears safe as a local text-processing skill: it reads user-selected text and writes cleaned output without contacting external services. Before installing, review the scripts, run the tool on copies first, and be careful with the batch in-place command. Also consider whether using it to bypass AI detectors is allowed in your context.
Findings (3)
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.
If run without backups or review, many text files could be changed at once.
The documented batch workflow can rewrite every .txt file in the current directory. This is user-directed and purpose-aligned, but it is still broad local file mutation.
for f in *.txt; do python scripts/humanize.py "$f" -o "$f.tmp" && mv "$f.tmp" "$f" done
Run on copies or version-controlled files first, review diffs, and avoid broad globs in directories containing important documents.
Users have less external context for who maintains the skill or where updates come from.
The registry information does not provide an upstream source or homepage for provenance checks. The provided code is simple and self-contained, so this is a provenance note rather than a concern.
Source: unknown Homepage: none
Review the included scripts before use and prefer pinned, auditable sources for future updates.
Using this skill may violate academic, workplace, publishing, or platform rules about AI-generated content disclosure.
The skill explicitly advertises detector-bypass use. This is disclosed and matches the implementation, but it can be used to misrepresent AI-assisted writing.
Use when you want to analyze text in AI detector and bypass it in future
Use it for legitimate editing and clarity improvements, and follow any rules requiring disclosure of AI assistance.