Back to skill

Security audit

帮你写大众点评的Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill only provides writing guidance for restaurant-review copy and does not install code, run commands, persist data, or access accounts.

Installers should expect the skill to analyze restaurant or dish photos and ask for location, environment, service, shortcomings, and use case details before drafting review text. Review generated claims before posting publicly, especially inferred price, restaurant type, or service details.

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
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
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description uses broad trigger phrases such as “写点评”, “帮我评价”, and “生成点评文案”, which can overlap with ordinary conversational requests and cause the skill to activate outside its intended context. While this does not directly create code execution or data exfiltration risk, ambiguous activation can route unrelated user inputs into this skill, increasing the chance of incorrect handling or unintended disclosure of user-provided images and context.

Content

No source excerpt is available for this finding.