Teller MCP – Borrow USDC & Altcoins (no margin calls)

• Outcome: Pass
• Latest audit: May 28, 2026, 11:49 AM
• Version:
  0.2.2

Overview

This is a coherent Teller MCP API wrapper, but users should manually review any generated borrow or repay transaction before signing it.

Install this only if you want an agent to query Teller lending data and prepare DeFi transaction payloads. Keep TELLER_API_BASE_URL pointed at a trusted endpoint, treat wallet and loan queries as financial metadata, and never sign generated borrow/repay transactions until you verify chain ID, contract addresses, approval spenders, amounts, and loan terms yourself.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 91% confidence
• Finding: The skill explicitly instructs users to install and run a bundled MCP server that uses environment variables and outbound network access, but the skill does not declare permissions for those capabilities. This creates a transparency and policy gap: operators or agents may execute code with broader access than the skill metadata communicates, increasing the chance of unsafe deployment or accidental data exposure.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The documentation explicitly instructs users to use generated borrow and repay transaction payloads directly as calldata for sequential on-chain submission, but it provides no warning that blockchain transactions are financially consequential, may be irreversible, and can approve token spending or move funds. In an agent skill context, this omission increases the chance that an automated system or user will treat the output as routine API data rather than sensitive transaction instructions requiring confirmation and validation.

Missing User Warnings

Medium

• Confidence: 86% confidence
• Finding: The README explicitly advertises wallet-specific loan retrieval and generation of encoded on-chain borrow/repay transactions, but it does not warn users that supplying wallet addresses exposes financial position data or that blindly submitting generated transactions can lead to loss of funds if parameters, chain, or approvals are incorrect. In an agent-integrated MCP context, missing safety guidance is more dangerous because downstream agents may treat the server output as trustworthy and automate privacy-sensitive queries or transaction execution without adequate human review.

VirusTotal

66/66 vendors flagged this skill as clean.