ClawHub
Back to skill

Security audit

CP skill test 123456

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward Xiaohongshu publishing helper, but it can click the real publish button without a clear final confirmation and may activate on drafting-style requests.

Review carefully before installing. This skill is not deceptive and has no executable payload, but it can operate a real Xiaohongshu publishing flow. Use it only when you intend to publish, and require the agent to ask for final confirmation before clicking publish.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger phrases include very common requests such as '写一篇小红书' and '发小红书', which can match ordinary drafting or brainstorming intents rather than an explicit request to publish externally. In a skill that drives a browser to a real publishing platform, overly broad activation increases the chance of unintended invocation and accidental posting workflows.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill walks through opening the publishing page and clicking the final '发布' button without any warning that this performs a real external post that may be difficult to undo or could have reputational consequences. In this context, the absence of an explicit confirmation gate makes accidental publication materially more dangerous than a normal browser automation step.

Static analysis

No suspicious patterns detected.