ClawHub
Back to skill

Security audit

数据分析与反馈

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a straightforward ad analytics helper, with the main caution that campaign performance data may appear in reports, events, and logs.

Use this with campaign analytics data you are comfortable exposing to the OpenClaw runtime, reports, events, and logs. Avoid feeding unnecessary personal data, verify who can view logs and exported CSV reports, and be careful opening CSV exports in spreadsheets because material IDs are not visibly escaped or formula-neutralized.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill explicitly handles advertising analytics data collection, analysis, and reporting, but provides no guidance on data sensitivity, access control, retention, minimization, or privacy-safe handling. In a real deployment this can lead to over-collection or unsafe processing of campaign or user-related data, especially because the documented API accepts arbitrary metric lists and raw data arrays without any stated constraints.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.