ClawHub
Back to skill

Security audit

Gemini Web Search

Security checks across malware telemetry and agentic risk

Overview

The skill is a plausible Gemini web-search helper, but its shell wrapper handles search text unsafely enough that a crafted prompt could run local commands.

Install only if you are comfortable with an agent sending relevant prompts to your local Gemini CLI and using that CLI's configured account. Avoid using the bundled shell wrapper or raw shell templates with untrusted prompt text until the command is changed to pass arguments without a shell.

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 (1)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill description uses very broad trigger phrases like 'what’s the latest news', 'search the web', and 'find sources/links', which can match a wide range of ordinary user requests and cause the skill to be invoked more often than intended. In an agent setting, ambiguous routing increases the chance that this skill handles tasks outside its narrow purpose, potentially leading to unnecessary external queries, privacy leakage of user prompts to a third-party CLI, or reduced control over tool selection.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.