Humanizer AI
ReviewAudited by ClawScan on May 10, 2026.
Overview
The package appears to do what it advertises, but its advertised purpose is to evade AI-detection systems and it sends submitted text to an external HumanizerAI API.
Review this carefully before installing. The code shown is a straightforward CLI wrapper, but the skill is specifically designed to bypass AI detectors and can send your text to an external service while using your API credits. Use it only for permitted rewriting tasks, avoid sensitive documents, verify the API endpoint, and require explicit approval before any batch processing or publication workflow.
Findings (4)
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.
Using this skill may help produce content that misleads reviewers, schools, employers, platforms, or readers about whether text was AI-generated.
The skill's core instruction is to make AI-generated text evade third-party AI-detection and integrity systems.
Detect AI patterns and rewrite text to bypass GPTZero, Turnitin, Originality.ai, and other detectors.
Only use it where this kind of rewriting is permitted, and require explicit user approval before using it to bypass detectors or publish content.
An agent could process more text than intended or spend HumanizerAI credits if a batch or large file is used.
The agent is allowed to run the scoped CLI, including credit-consuming and batch file workflows; this is disclosed and central to the skill, but users should notice the automation potential.
allowed-tools: Bash(humanizerai:*) ... Humanization (1 credit = 1 word) ... Pattern 2: Batch Humanize Multiple Files
Confirm the exact files, word count, intensity, and credit impact before allowing batch or large humanization jobs.
Anyone or any agent with access to the environment variable can use the account's HumanizerAI API allowance.
The CLI uses the HumanizerAI API key as a bearer token for service requests, which is expected for this integration but gives access to the user's HumanizerAI account and credits.
Authorization: `Bearer ${this.apiKey}`Store the API key securely, restrict who can run the skill, and rotate the key if it may have been exposed.
Private drafts, essays, customer text, or confidential documents may be transmitted to the provider if supplied to the CLI.
User-provided text is sent to the external HumanizerAI API for detection or rewriting; this is purpose-aligned, but it creates a data-sharing boundary.
this.apiUrl = config.apiUrl || 'https://humanizerai.com/api/v1'; ... body: JSON.stringify({ text, intensity })Do not submit sensitive or confidential text unless the provider's privacy terms are acceptable, and verify any HUMANIZERAI_API_URL value points to the intended endpoint.