ClawHub
Back to skill

Security audit

探店脚本台

Security checks for vulnerabilities and agentic risk

Overview

This is a simple restaurant store-visit video script-writing skill with no code execution, credential access, network use, or persistence.

Installers should understand this skill is for restaurant/store-visit promotional copy. Review generated claims for accuracy, disclose paid or cooperative promotions as required by the target platform, and avoid using unauthorized music or materials.

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 (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger keywords include very generic phrases such as ‘拍店’ and ‘餐饮推广’, which can overlap with ordinary user requests and cause the skill to activate outside its intended scope. Over-broad activation can route unrelated conversations into this skill, leading to prompt hijacking of task selection, degraded assistant behavior, or inappropriate use of the skill in contexts where it was not requested.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The natural-language trigger examples are broad and underspecified, such as requests to ‘整理成一张表’ or ‘改成更简短的版本’, which could match many unrelated conversations once partial context exists. This increases the chance of accidental invocation or persistence of the skill beyond its intended boundary, potentially confusing workflow routing and causing unintended content generation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.