Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Humanizer Enhanced
v1.0.0Advanced AI text humanizer for blog content. Detects and removes 34 AI writing patterns, adds personality/soul, and handles crypto/Web3 specific tells. Use w...
⭐ 0· 1.2k·4 current·4 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description align with the content of SKILL.md and the three reference docs: it scans for 34 AI-writing patterns, reports severity, optionally edits files, and injects 'soul' and voice. No unrelated permissions, binaries, or credentials are requested.
Instruction Scope
Instructions tell the agent to scan and modify local files (single file or batch), show results, and ask for approval before fixing — this is within scope. Two points to watch: (1) the references push replacing vague attributions with specific sources (e.g., 'according to DeFiLlama' or named studies). Because the skill has no mechanism to verify external facts, the agent may be given broad discretion to insert or invent citations unless callers constrain it. (2) 'Add soul/personality' guidance is intentionally open-ended and grants stylistic latitude; review is recommended before publishing.
Install Mechanism
No install spec or code is included (instruction-only). Nothing will be written to disk by an installer and no third-party packages are pulled in by the skill itself.
Credentials
The skill requires no environment variables, no credentials, and no config paths. Its file operations (scan/edit) are consistent with editing local content and are proportionate to the described capability.
Persistence & Privilege
always:false (default); the skill does not request permanent inclusion or special privileges and does not modify other skills' configurations. Autonomous invocation is allowed by platform default but is not elevated by the skill itself.
Assessment
This skill is internally consistent and lightweight: it will scan and (with permission) edit local text files to remove AI-writing patterns and add personality. Before installing or running it, note: (1) it can batch-edit files — always preview changes and keep backups; (2) the guidance encourages replacing vague claims with specific citations; if the agent cannot verify sources, it may insert inaccurate or fabricated attributions, so require verification or manual citation review; (3) stylistic 'soul' edits are subjective — review edits for tone and accuracy, especially for crypto/Web3 facts or claims. No credentials or installs are needed, so the primary operational risk is accidental overwriting of content and potential insertion of unverifiable facts rather than credential exfiltration.Like a lobster shell, security has layers — review code before you run it.
latestvk971wsbgqvkdyj949qy01cw0f981ea7y
License
MIT-0
Free to use, modify, and redistribute. No attribution required.