ClawHub

Cloudflare Crawl

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Cloudflare crawling helper that uses user-provided Cloudflare credentials to start and manage crawl jobs.

Install only if you intend to use Cloudflare Browser Rendering for website crawling. Use a least-privilege Cloudflare API token, set modest limits and depths, avoid crawling private or sensitive pages unless you can protect the output, and be careful with --verbose or --json because those can print crawled content into terminal logs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill requires environment secrets and clearly performs outbound network operations, but it does not declare explicit permissions for those capabilities. That creates a transparency and governance gap: users or platforms may not realize the skill can access credentials and send data to external services, increasing the risk of unintended secret use or unreviewed data egress.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The skill enables potentially large-scale crawling and even shows an example that writes crawled content to local files, but it does not warn users about resource consumption, storage growth, target-site policy considerations, or the implications of bulk data collection. In practice, this can lead to unexpected external requests and unbounded local artifact creation during use.

Missing User Warnings

Medium
Confidence
80% confidence
Finding
The tool can print crawled page content previews and full raw JSON results directly to stdout, which may expose sensitive information from scraped pages, terminal logs, CI logs, or shared shell history. In a crawling/extraction context, this risk is elevated because the operator may target authenticated or internal content and the output handling has no redaction or warning controls.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal