Back to skill

Security audit

aigc-reducer

Security checks for vulnerabilities and agentic risk

Overview

This text-only skill does not appear to access your system, but its main purpose is to disguise AI-written academic text from detectors like Turnitin.

Review carefully before installing. The package is not showing system-access or malware behavior, but it is designed to help conceal AI-generated academic writing from detection systems, which may violate school, publisher, or workplace integrity rules.

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 (6)

Natural-Language Policy Violations

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill description explicitly promotes bypassing AIGC and Turnitin detection, which is a direct evasion objective rather than a neutral writing aid. This enables academic-integrity abuse and deliberate concealment of AI-generated authorship, making the context substantially more dangerous than ordinary paraphrasing or copyediting.

Content

No source excerpt is available for this finding.

Ssd 2

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The top-level framing states that the skill should reduce AI-detection rates and make text look human-written, which is a semantic evasion pattern. Because the core function is concealment of origin, the skill facilitates deceptive transformation of content to defeat text-analysis systems.

Content

No source excerpt is available for this finding.

Ssd 2

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The pipeline explicitly sequences modules to remove machine-like traits and inject human-like stylistic variation while preserving the underlying content. That combination is dangerous because it operationalizes stealthy paraphrasing as a repeatable process for defeating AI-authorship detection at scale.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger conditions are unusually broad and include ordinary user intents, which increases the chance the skill is invoked outside narrow, legitimate editorial contexts. In this specific skill, that breadth is more dangerous because the advertised purpose is evading AI-authorship and Turnitin detection, so over-triggering can route benign writing requests into an evasion workflow.

Content

No source excerpt is available for this finding.

Ssd 2

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

These module instructions emphasize indirect linguistic tactics such as obscuring connective patterns, softening certainty, and increasing perplexity to avoid recognizable generation signals. In context, these are not ordinary style suggestions but components of an evasion strategy intended to make machine-generated text harder for analysis systems to identify.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The document title and heading present a prescriptive Chinese-only writing standard for identifying and replacing so-called AI jargon, including the directive '必须替换'. This imposes a language/locale-specific policy without any opt-in, alternative language support, or justification that the skill is limited to a Chinese-language context.

Content

No source excerpt is available for this finding.