Trade Executor

Security checks across malware telemetry and agentic risk

Overview

This skill openly enables live crypto trading and includes confirmation and risk-control instructions, but it should only be installed by users who intentionally want an agent able to place real exchange orders.

Install only if you deliberately want agent-assisted live crypto trading. Use exchange API keys restricted to trading, disable withdrawals and transfers, keep exchange-side limits small, verify every order preview before confirming, and understand where gateway audit logs and risk-tracking data are stored.

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)

Natural-Language Policy Violations

Medium

Confidence: 89% confidence
Finding: Requiring an exact Chinese reply ("确认" / "取消") for a security-critical approval step creates an unsafe confirmation mechanism. Users or upstream agents may misunderstand the required response, and brittle exact-match logic can lead to failed cancellations, accidental approvals through poor normalization, or pressure to bypass confirmation handling in automation; in a trade-execution skill, ambiguity in consent directly affects financial safety.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: This implementation document provides concrete instructions for placing authenticated live Binance orders and OCO orders using real API credentials, but it does not include explicit safeguards or warnings that these actions move real funds. In the context of a trade-execution skill, omission of prominent confirmation, simulation/testnet guidance, and account-impact warnings increases the risk of unintended real-money trades by an agent or user.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: This implementation documents live order placement, order status checks, and stop-loss/take-profit execution against a real exchange, but it does not include an explicit warning that these actions will execute real trades affecting user assets. In an agent skill context, that omission is dangerous because an LLM-driven agent may treat the instructions as ordinary API usage and initiate irreversible financial transactions without sufficiently prominent user awareness and consent.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal