The Agent Skills Directory

[Data Exposure & Exfiltration] (SAFE): The skill does not access sensitive local file paths or attempt to send data to external servers. It contains one informational URL (gptcleanup.eu) for manual use, which is not called programmatically.
[Obfuscation] (SAFE): No obfuscation (Base64, zero-width characters) was detected in the skill's code or instructions. The skill explicitly includes logic to detect and remove Unicode watermarks from processed text as a security and quality feature.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): No external packages (npm/pip) are requested. The provided Python snippet for watermark removal is local string manipulation and does not involve remote code execution or shell commands.
[Indirect Prompt Injection] (SAFE): While the skill processes untrusted user-provided text, it lacks any dangerous capabilities (e.g., file writing, network requests, or system command execution) that could be exploited via indirect injection. It acts as a closed-loop text transformer.
[Privilege Escalation & Persistence] (SAFE): The skill does not attempt to acquire elevated permissions or establish persistence mechanisms on the host system.

ukrainianizer