Humanizer Enhanced
ReviewAudited by ClawScan on May 1, 2026.
Overview
This is an instruction-only writing editor with no code or credentials, but it can rewrite batches of content and is designed to make AI-assisted text look more human, so edits should be reviewed carefully.
This skill appears safe to install from the provided artifacts because it is instruction-only and does not request code execution, credentials, or network access. Before using it, decide whether making AI-assisted text sound human is appropriate for your publication context, run scan-only mode when possible, and review all edits for accuracy, disclosure, and tone.
Findings (2)
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.
Published readers may assume the text was written without AI assistance, and factual caveats could be softened during rewriting.
The skill's disclosed purpose is to remove AI-writing signals and make content appear human-authored, which is purpose-aligned but may affect reader trust or disclosure obligations.
Identify and remove signs of AI-generated text... adds personality to make content sound genuinely human-written.
Use the skill transparently, preserve required AI-use disclosures, and verify any claims, sources, or statistics before publishing.
A broad batch run could unintentionally alter multiple drafts if the user does not review the proposed edits.
The skill can rewrite multiple Markdown files in a directory, but the artifacts show this as a user-invoked workflow with an approval step before fixing.
/humanizer --batch drafts/ # Process all .md files in directory ... Ask user: "Fix all issues? Or review one by one?"
Run scan-only mode first, keep backups or use version control, and review diffs before accepting batch edits.