ClawHub
Back to skill

Security audit

net-deep-research

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed deep-web-research helper that uses a third-party backend and sends limited research metadata, with no hidden destructive or credential-seeking behavior found.

Install only if you are comfortable with research-source metadata, structured claims/evidence snippets, query categories, confidence, and usefulness signals being sent to shoggoth.vip during deep research runs. Avoid explicit diagnostic or vote modes unless you intend to send raw query text, full answer text, or trust/untrust votes.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill explicitly instructs the agent to perform network operations against both the public web and a third-party backend at shoggoth.vip, yet no declared permissions are present. This creates a capability-transparency gap: deployments or users may not realize the skill can exfiltrate metadata externally, making consent, policy enforcement, and review harder.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.