ClawHub
Back to skill

Security audit

reg-citation-normalizer

Security checks for vulnerabilities and agentic risk

Overview

This skill is an offline citation-normalization helper that reads user-supplied regulatory text or files and does not show hidden network, persistence, privilege, or destructive behavior.

Install this if you want an offline helper for regulatory citation formatting. Be aware that check mode reads the document path you provide and may echo short line contexts in its output; do not treat its normalized citations as proof that a regulation or clause is current or legally correct.

A.I.G

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

SkillSpector

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 (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The English trigger list includes broad terms such as regulatory citation names and generic normalization phrases that could be mentioned in ordinary conversation, causing the skill to activate outside a clearly intentional invocation. Over-broad activation can lead to unintended context capture, tool selection, or workflow hijacking in agent environments even though the skill itself appears non-executive and offline.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The Chinese trigger list uses broad regulatory terms like '法规引用' and standard names that may appear in normal drafting or discussion, making accidental activation plausible. In an agent system, such ambiguous triggers can cause the wrong skill to run and influence outputs or processing flow without the user explicitly requesting citation normalization.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The file’s primary descriptive content is presented exclusively in Chinese, which can impose a language constraint on users or agents without explicit opt-in. The policy allows locale-specific behavior when it is justified or optional, but this file does not state that the skill is Chinese-only or offer an alternative language.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.