ClawHub

Gate VIP & Fee Query Skill

Security checks across malware telemetry and agentic risk

Overview

The skill is a read-only Gate account fee lookup, but it makes mutable remote runtime rules authoritative over agent behavior.

Review before installing because the skill delegates mandatory runtime behavior to an unpinned remote GitHub rules file. Use only Gate OAuth or a read-only Gate API key with the documented Account:Read and Wallet:Read permissions, and avoid granting trading or withdrawal permissions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The scenario instructs the agent to retrieve account profile and fee data, which are user-specific financial details, without requiring an explicit notice or consent step before accessing them. In a conversational agent context, silent retrieval of sensitive account information can violate user expectations, reduce transparency, and increase the chance of unintended disclosure if the request was ambiguous or made in a shared environment.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The combined-query flow aggregates data from multiple account endpoints but does not tell the user that more than one sensitive backend call will be made. This lack of transparency is more dangerous than the single-endpoint cases because it expands the scope of accessed account data, increasing privacy risk and the likelihood of over-collection beyond what the user expected.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal