Humanizer
PassAudited by ClawScan on May 13, 2026.
Overview
This is a coherent instruction-only writing/editing skill with no code, install step, credentials, or network access, but it can read and edit user-specified local files and load an optional project context file.
This skill appears safe for its stated purpose. Before installing, know that it can read and edit local text files you point it at, and it will use a project-level humanizer-context.md file if one exists. Keep sensitive information out of that context file and review changes before publishing or committing edited documents.
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.
If used with --mode edit and --file, the agent may alter the selected document in place.
The skill is allowed to read and modify files. This is aligned with its documented file-editing purpose, but users should understand that local files can be changed.
allowed-tools: - Read - Write - Edit - Grep - Glob - AskUserQuestion
Use detect or rewrite mode first for important documents, keep backups, and provide only the file paths you intend to edit.
Brand samples, banned phrases, or other content in humanizer-context.md may affect outputs and be exposed to the model session.
The skill can automatically read a persistent local context file and use it to guide rewriting. This is purpose-aligned, but the file contents can influence model behavior and may be included in the agent's working context.
Auto-loads `humanizer-context.md` from the project root if present. Use that file for brand samples and banned phrases.
Review humanizer-context.md before use and avoid placing secrets, credentials, or sensitive private material in that file.