Humaniseur Fr

• Outcome: Pass
• Latest audit: May 28, 2026, 2:55 PM
• Version:
  1.0.3

Overview

This is a French text-rewriting skill with no executable code or hidden data behavior, though its activation wording and tool metadata are broader than necessary.

Safe to install for French writing cleanup. Be aware it may activate on broad humanizing or rewriting requests, and review the broad tool list and mismatched crypto/purchase capability tags if you prefer tightly scoped writing skills.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The activation text uses very broad phrases such as editing, reviewing, rewriting, cleaning up, and generic 'make it sound human' triggers. In an agentic environment, this can cause the skill to activate on ordinary writing requests and silently steer outputs toward its own transformation rules, overriding user intent or conflicting with other skills.

Natural-Language Policy Violations

Medium

• Confidence: 88% confidence
• Finding: The instruction set mandates French output behavior and preservation of a French register without checking whether the user asked for French or whether the surrounding task permits a language shift. If auto-invoked, this can cause instruction interference, unexpected language coercion, and degraded reliability in mixed-language or non-French workflows.

VirusTotal

65/65 vendors flagged this skill as clean.