Back to skill

Security audit

Humanize Ai Text

Security checks for vulnerabilities and agentic risk

Overview

This skill is a low-risk writing helper, though its package appears incomplete because it documents scripts that are not included.

Before installing, understand that this appears to be a writing-assistance skill rather than a working local detector package as published: the referenced Python scripts are missing. Review any rewritten text manually, especially where authorship transparency matters.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The natural-language description and usage examples are entirely in Chinese, and the feature list explicitly marks multilingual support as unavailable. This effectively imposes a language constraint without documenting user opt-in or a justified locale-specific limitation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The README provides example phrases for using the skill, but it does not define explicit activation boundaries, exclusions, or context. Phrases like '把这段话改得像人类写的' are broad rewriting requests that could overlap with ordinary user speech, making unintended invocation more likely.

Content

No source excerpt is available for this finding.