Humanizer (Polish)
PassAudited by ClawScan on May 1, 2026.
Overview
This appears to be a benign, instruction-only Polish editing skill, but it can read/edit documents and is designed to make AI-written text sound more human.
Before installing, understand that this is an instruction-only text humanizer with local file editing permissions. Use it on specific documents or pasted text, review edits before saving, and use it responsibly where AI-authorship disclosure rules apply.
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.
The agent could change a document while humanizing text, so accidental edits to the wrong file are the main practical risk.
The skill can read, search, write, and edit local files. That fits document editing, but it gives the agent file-mutation capability if invoked on files.
allowed-tools:\n - Read\n - Write\n - Edit\n - Grep\n - Glob\n - AskUserQuestion
Specify the exact text or file to edit, ask for a diff or preview before saving, and keep backups for important documents.
It is harder to independently verify the package origin or confirm which version the registry is presenting, though no executable code or install script is present.
The registry metadata has limited provenance information, and the visible SKILL.md frontmatter shows a different version value, creating a minor packaging/versioning ambiguity.
Source: unknown; Homepage: none; Version: 0.1.0
Install only from the trusted registry entry, check the owner/version before updates, and prefer packages with clearer source provenance when available.
Using it may make readers believe a text was written more directly by a human than it was.
The skill is explicitly intended to make AI-written text appear more natural and human, which is transparent but can affect downstream trust in authorship.
usuwa ślady pisania AI z tekstu polskiego, nadając mu bardziej naturalny, ludzki charakter
Use it for legitimate editing and clarity, and follow any disclosure, academic, workplace, or platform rules about AI-assisted writing.