ClawHub
Back to skill

Security audit

snaptrade-API

Security checks across malware telemetry and agentic risk

Overview

This SnapTrade skill is transparent and purpose-aligned, but it gives an agent real brokerage trading and cancellation authority with broad routing and uneven confirmation guidance.

Install only if you intentionally want OpenClaw to access brokerage data and potentially place or cancel real orders through SnapTrade. Prefer paper trading or a low-limit dedicated account first, keep credentials in a secret manager or private .env file, and require explicit confirmation of account, symbol, side, quantity, order type, price, and cancellation target before any live action.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill clearly accesses environment variables containing highly sensitive brokerage credentials and trading authority, yet no explicit permissions are declared. That weakens platform governance and user visibility into what the skill can access, increasing the chance of over-privileged deployment and accidental secret exposure through downstream code paths.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger text is extremely broad ('always use this skill when brokerage account interaction of any kind is needed'), which can cause unintended invocation for ambiguous user requests. In a trading context, mistaken routing is especially dangerous because the skill can access account data and place or cancel real orders.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
This documentation gives straightforward instructions for placing real crypto orders, including direct use of `place_crypto_order`, but it does not require an explicit confirmation, sandbox mode, or a prominent warning that the action will execute a live trade. In a trading skill that is triggered on natural-language requests like 'buy X shares,' this increases the risk of unintended or ambiguous user prompts resulting in irreversible financial transactions.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
This documentation provides direct instructions for placing real option orders, including multi-leg trades, without an explicit warning that these actions can execute against a live brokerage account and materially change funds, margin exposure, and positions. In the context of an agent skill that is triggered for brokerage interactions, omission of a clear confirmation/warning step increases the risk of unintended or overly broad autonomous trade execution.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.