LI.FI MCP Skill

• Outcome: Pass
• Latest audit: May 29, 2026, 8:22 PM
• Version:
  1.0.0

Overview

This skill is a disclosed LI.FI route-planning helper that uses UXC to query a LI.FI MCP endpoint and does not include hidden execution, signing, broadcasting, or data exfiltration behavior.

Install only if you trust your local UXC setup and are comfortable sending wallet addresses, transaction hashes, route details, and token choices to LI.FI. Treat returned quotes and transaction requests as informational until you independently verify chain, token, spender, recipient, amount, and fees in your wallet before signing externally.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence
• Finding: This is a mismatch because the provided code does not interact with the LI.FI MCP server, perform route discovery, bridge/DEX checks, quote generation, balance/gas/allowance checks, token or chain lookup, or transfer tracking. Instead, it only validates that documentation files exist and contain specific strings and formatting. Its primary purpose is documentation/config linting for the skill, which is materially different from the declared runtime functionality.

VirusTotal

65/65 vendors flagged this skill as clean.