ClawHub
Back to skill

Security audit

Realtime Web Search

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent Baidu-backed realtime web search tool, with normal privacy cautions for sending and logging search queries.

Install only if you are comfortable sending search queries to Baidu with your BAIDU_API_KEY. Avoid secrets or private data in queries, use a limited-scope key, keep endpoint override variables pointed only at trusted Baidu-compatible endpoints, and remember that query text may appear in local logs because of the debug print.

SkillSpector

By NVIDIA
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
Confidence
78% confidence
Finding
The description advertises very broad trigger phrases such as `查一下/搜一下/最新消息/实时信息`, which can match ordinary user requests and cause the skill to activate more often than intended. Over-broad activation increases the chance of unintended external web queries, unnecessary disclosure of user prompts to third-party services, and accidental reliance on live web results when the user did not explicitly request tool use.

Vague Triggers

Low
Confidence
74% confidence
Finding
The common use case list repeats vague triggers like `查一下 / 搜一下 / 最新消息` without defining clear activation boundaries or exclusions. This reinforces permissive routing behavior and may cause the agent to overuse the skill for generic questions, expanding unnecessary network exposure and third-party data sharing.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal