ClawHub
Back to skill

Security audit

实时财经

Security checks across malware telemetry and agentic risk

Overview

This finance-news skill is not clearly malicious, but it should be reviewed because it advertises API-only fetching while bundling a helper that can attach to a local browser debugging session.

Install only after reviewing the scripts. Prefer the documented fetch_api.js path, avoid running fetch_and_save.js unless you intentionally started a dedicated browser debugging session for this skill, and make sure any 5-minute scheduled polling is something you can disable. The publisher should either remove the browser helper or clearly document and gate it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Tp4

High
Category
MCP Tool Poisoning
Confidence
85% confidence
Finding
The skill description presents a narrow API-based news fetcher, but the documented/observed behavior includes persistent local storage, state tracking, database-style management, and especially Puppeteer access to a locally exposed browser debugging port. That mismatch is security-relevant because users or orchestrators may grant the skill broader trust than intended, while browser-debug access can expose unrelated open pages, session data, or sensitive content beyond the stated finance-news task.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal