ClawHub
Back to skill

Security audit

Full Potential Intelligence

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed remote MCP skill for AI intelligence tools, with reasonable cautions around trusting the service and avoiding sensitive submissions.

Install only if you trust fullpotential.ai and the mcp-remote connector. Treat prompts, tool arguments, field reports, registration details, and credit-spending requests as data sent to a third-party service; avoid confidential content and confirm before write or metered actions.

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
95% confidence
Finding
The skill instructs users to connect an MCP client to a third-party remote endpoint over SSE via `npx mcp-remote` without any disclosure about what prompts, tool arguments, usage metadata, or other workspace-derived data may be transmitted off-platform. Because MCP tools can be invoked within an agent workflow, this creates a real data exposure and trust-boundary risk, especially when the skill markets write-capable and metered services but does not warn users to review privacy, authentication, or data-handling implications before connecting.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.