ClawHub
Back to skill

Security audit

崩溃与性能数据查询

Security checks across malware telemetry and agentic risk

Overview

The skill appears to query Umeng app analytics, but it also includes under-disclosed telemetry that may send app-specific identifiers outside the requested analysis workflow.

Review this skill before installing. Only use it if you are comfortable with possible telemetry in addition to Umeng API queries, and avoid using sensitive production app identifiers unless the publisher documents exactly what is sent, where it goes, and how to disable it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The skill is presented as supporting 8 read-only U-APM query APIs, but it also instructs the agent to run `umeng-cli trace` telemetry commands that are outside that stated scope. This creates hidden side effects and causes external data transmission unrelated to the user's requested APM query, violating least surprise and expanding the skill's effective behavior.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill directs the agent to perform telemetry reporting without informing the user that skill usage metadata, and potentially an appKey, will be sent to an external service. Undisclosed exfiltration of task metadata is a security and privacy issue because it leaks information beyond what is required to fulfill the user's request.

Ssd 3

Medium
Confidence
98% confidence
Finding
The skill explicitly tells the agent to capture a user-provided Appkey and include it in a telemetry trace command before making the first API call. Even if an appKey is not always a secret in the strictest sense, it is application-specific identifier data and may be sensitive in enterprise contexts; transmitting it to a third party without need or consent materially increases data exposure risk.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.