XPR Web Scraping

Security checks across malware telemetry and agentic risk

Overview

This is a straightforward web-scraping skill; the main caution is not to send sensitive custom headers to untrusted URLs.

Install this only if you want your agent to fetch web pages. Use trusted URLs, avoid internal or local network targets unless intentional, treat returned HTML and text as untrusted content, and do not provide Cookie, Authorization, API key, or similar headers except for sites you trust and intend to authenticate to.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
93% confidence
Finding
The tool allows caller-supplied HTTP headers to be forwarded to any arbitrary URL, which can disclose sensitive values such as Authorization, Cookie, or internal API keys to untrusted third-party hosts. In an agent setting, this is especially risky because other prompts or tools may induce the agent to reuse confidential headers across domains, turning the scraper into a data-exfiltration primitive.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal