ClawHub
Back to skill

Security audit

像素级凝视

Security checks for vulnerabilities and agentic risk

Overview

This is a prompt-only image inspection skill with no code execution, persistence, credential use, or hidden data movement.

Installers should expect this skill to make image and screenshot reviews more deliberate and sometimes more verbose. Users who do not want Chinese-language examples or broad activation on visual tasks may want the publisher to clarify language behavior and narrower trigger conditions, but no security-blocking behavior was found.

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 (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill metadata and presentation strongly bias the interaction toward Chinese without any user-language negotiation. For users operating in another language, this can degrade usability, produce confusing outputs, and cause the skill to override the agent's normal language alignment with the user's request. The issue is less dangerous than code-execution or data-exfiltration flaws, but it is still a prompt-behavior vulnerability because it can disrupt expected assistant behavior at activation time.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The activation description is very broad: it triggers on almost any image-analysis, OCR, comparison, or screenshot-reading request. Over-broad activation can cause the skill to intercept unrelated user tasks and override normal response behavior, increasing the chance of inappropriate skill invocation and prompt-scope expansion. In this case the content is not overtly malicious, but it is broad enough to affect routing and behavior across many common requests.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The entire skill file is written in Chinese and provides no indication that users may choose another language or locale. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is clearly documented and justified.

Content

No source excerpt is available for this finding.