Back to skill

Security audit

应用市场监测

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AppGallery reporting workflow that reads user-provided app lists, queries public app-store information, and returns a PDF report.

Install this if you are comfortable with the agent reading the app names and optional update dates from attached Excel files when you use the stated monitoring keyword, operating AppGallery through a phone GUI, and returning a generated PDF report. Avoid attaching spreadsheets with unrelated sensitive columns even though the skill says they are ignored.

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 (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger phrases include broad, natural requests like '查一下这些 App 的更新情况' and '帮我查应用市场的版本', which can match ordinary user messages and cause the skill to activate unexpectedly. In this skill, unexpected activation is more concerning because it can drive a GUI agent and generate reports, increasing the chance of unintended automated actions on user-supplied data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The one-click mode automatically parses attached Excel files when the keyword appears, but the skill does not prominently warn users that attachments will be ingested and processed without further confirmation. This is dangerous because users may mention the keyword casually while sharing spreadsheets, leading to unintended processing of potentially sensitive business data and downstream report generation.

Content

No source excerpt is available for this finding.