Security audit

xcard trade

Security checks across malware telemetry and agentic risk

Overview

This is a transparent crypto futures trading skill, but it can access sensitive account history and perform live account changes with incomplete user-control safeguards.

Install only if you trust the publisher and can use a restricted XCard API key. Avoid keys with withdrawal permission, and require manual confirmation for every place, cancel, modify, leverage-change, or account-transaction-history request, especially anything involving deposits or withdrawals.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Description-Behavior Mismatch

Medium

Confidence: 91% confidence
Finding: The documented `/account/transactions` endpoint exposes deposits and withdrawals in addition to funding and trade events, which exceeds the stated skill scope of trading, positions, funding rates, and margin management. In a finance context, account transaction history is highly sensitive and can reveal wallet activity, balances, and user financial behavior without an explicit scope warning or narrower access boundary.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The trigger phrases are broad enough to activate on common finance-related words like "balance," "margin," or "account" without confirming the user's intent to use this high-risk trading skill. In a crypto futures context, unintended activation can expose sensitive account data or lead the agent into account-specific or trading workflows when the user may have only asked a general question.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: This file documents endpoints that return sensitive financial account information and require an API key, but it does not instruct the agent to protect credentials, avoid echoing secrets, or minimize disclosure of wallet and position data. In a trading skill, missing privacy and secret-handling guidance increases the chance of credential leakage or overexposure of balances, margin, and PnL to unintended recipients or logs.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger list includes broad terms like `history`, `trade`, `pnl`, and common multilingual words that can match ordinary conversation, increasing the chance that the skill is loaded when the user did not clearly request sensitive financial history. Because this module can access executed trades, closed positions, and account transactions, overbroad activation can lead to unnecessary exposure of private account data.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: This module handles sensitive financial history, including realized PnL, closed positions, and account transactions, but does not include a user warning or consent cue about accessing private historical account data. In a trading skill, the context makes the data especially sensitive because it can reveal investment behavior, losses, deposits, and withdrawals, so lack of notice increases privacy and overreach risk.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger list includes very broad generic terms like "price" and "quote," which can match many unrelated user requests and cause this trading skill to activate unexpectedly. In a crypto futures trading context, unintended activation is more dangerous because it can steer conversations toward leveraged financial actions or expose market/trading functionality when the user did not intend to invoke it.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.