ClawHub
Back to skill

Security audit

Gate Exchange Referral Skill

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly referral guidance, but it makes an unpinned external GitHub rules file highest-priority runtime instructions while also depending on Gate MCP setup that may involve exchange account authentication.

Review before installing. The referral content itself is mostly read-only and purpose-aligned, but users should be aware that the skill delegates runtime authority to an external mutable rules file and may lead them toward Gate MCP authentication. Only authorize Gate MCP or API credentials when you specifically need account-scoped Gate functionality, and prefer least-privileged credentials.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list is broad enough to match generic support and exchange-related language such as 'invite', 'referral', 'coupon', 'commission', and 'how to earn', which can cause the skill to activate outside its intended scope. Misrouting users into a promotional/referral skill can lead to incorrect guidance, overshadow more appropriate skills, and steer users toward referral flows when they were seeking neutral account or support information.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal