ClawHub

Liberfi Perpetuals

Security checks across malware telemetry and agentic risk

Overview

This is a coherent crypto trading skill, but it asks for persistent account access and a silent global CLI install while under-scoping consent for sensitive account reads.

Install only if you trust LiberFi and its CLI. Verify or install the CLI yourself, confirm which LiberFi account and wallet the agent is using before viewing private trading data, know how to revoke the agent login, and never approve deposits, order submissions, or cancellations unless the amount, recipient, market, side, size, and fees are clear.

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 (3)

Ssd 3

Medium
Confidence
93% confidence
Finding
These instructions tell the agent to silently authenticate, resolve the user's server-managed wallet address, and query positions/orders/fills without explicit disclosure or consent at runtime. That creates a hidden identity-resolution and account-access flow that can surprise users and expose sensitive financial account data tied to the authenticated session.

Ssd 3

Medium
Confidence
92% confidence
Finding
The skill repeats a hard requirement to auto-resolve the user's TEE EVM address and forbids asking the user, reinforcing a covert retrieval pattern for personally linked wallet identity. In a financial context, silently binding 'my positions' to a hidden server-managed wallet increases privacy and authorization risk if sessions are stale, shared, or misunderstood.

Ssd 3

Medium
Confidence
94% confidence
Finding
The step-by-step 'My...' autoflow operationalizes hidden login, identity lookup, and personal-history queries, making unauthorized or non-transparent access easier to reproduce. Because the queried data includes trading history and PnL-related information, misuse could disclose sensitive financial behavior to someone interacting through an already-authenticated agent session.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal