Longbridge Openapi

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Longbridge finance skill that can access brokerage data and create alerts or DCA plans, but its sensitive actions are purpose-aligned and guarded by explicit confirmation steps.

Install only if you intend to use Longbridge account and market workflows. Use the official Longbridge CLI/MCP, grant the least privilege needed, keep tokens out of shared environments, and carefully review every preview before confirming watchlist changes, price alerts, statement exports, or DCA plans.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The manifest description is extremely broad and enumerates a huge set of common finance-related phrases, including casual queries and account-related intents. That can cause the skill to be invoked in situations the user did not clearly intend, increasing the chance of exposing private brokerage data or steering the agent into high-risk finance workflows unnecessarily.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The skill supports sensitive account operations such as holdings, balances, cash flow, statements, and exports, but the privacy warning is buried later in the document rather than presented prominently up front. In an agent setting, this raises the risk that private financial data is retrieved or written to files before the user fully understands the sensitivity or visibility implications.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal