ClawHub
Back to skill

Security audit

Baidu

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only Baidu helper with disclosed local memory and clear limits around account-changing actions.

Install only if you want Baidu-specific routing and optional local memory in ~/baidu/. Review that folder periodically, do not store credentials or sensitive business data there, and approve any Baidu console, key, cloud-resource, or billing action explicitly before the agent proceeds.

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)

Natural-Language Policy Violations

Medium
Confidence
91% confidence
Finding
The template preconfigures a language preference and explicitly asks whether Chinese-first sources are allowed, which can steer future outputs toward a specific language and source ecosystem before the user has opted in. In a region-aware Baidu skill, this can cause the agent to default to Chinese-language materials or China-specific sources that the user may not understand or expect, increasing the risk of misunderstanding, inaccessible guidance, or inappropriate source selection for sensitive tasks.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.