ClawHub
Back to skill

Security audit

clawhub-skill-remote-agent

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward remote-agent bridge that sends user-provided queries to a configured endpoint, with privacy and TLS cautions but no hidden or destructive behavior found.

Install this only for remote agents you control or trust. Treat any query sent through it as shared with that remote service, use a scoped and revocable REMOTE_AGENT_KEY if authentication is needed, verify REMOTE_AGENT_URL before use, and avoid --insecure except in controlled testing with non-sensitive data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill instructs use of a Python client that depends on environment variables and sends requests to an external HTTP endpoint, but the skill metadata does not declare corresponding permissions. This creates a transparency and governance gap: operators and users may not realize the skill can access environment-provided secrets and perform network egress to third-party systems.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The activation guidance is broad enough to trigger on many ordinary requests whenever a task seems 'domain-specific' or handled by a separate agent system. In context, broad routing increases the chance that sensitive user prompts, enterprise questions, or operational commands are forwarded externally without sufficient user awareness or necessity.

Missing User Warnings

High
Confidence
97% confidence
Finding
The skill describes forwarding user queries to an external HTTP service but does not present a clear warning to the user that their input may leave the local system and be processed by a third party. This is especially dangerous because the examples include enterprise knowledge, financial analysis, and operational requests, which may contain confidential or regulated information.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.