ClawHub
Back to skill

Security audit

content-sanitization

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only skill that gives prompt-injection sanitization guidance and does not install code, request credentials, or perform actions.

Before installing, understand that this skill is guidance, not the bundled sanitizer hook it references from the upstream plugin. Its broad activation wording may make it appear in many external-content workflows, but the reviewed package does not perform actions on its own.

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 (1)

Vague Triggers

Medium
Confidence
97% confidence
Finding
The trigger `or any untrusted input` is extremely broad and could match many ordinary situations beyond this skill's intended scope. Although the document later narrows usage, the manifest trigger itself lacks concrete constraints or negative examples, creating ambiguity about when the skill should activate.

Static analysis

No suspicious patterns detected.