Back to skill

Security audit

xcard trade

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed XCard crypto futures trading skill, but it grants high-impact trading and account-history access with several under-scoped safety controls.

Review this skill carefully before installing. Use a tightly scoped XCard API key if available, avoid withdrawal permissions, and expect the agent to access sensitive balances, positions, trade history, and transaction history. Require the agent to show exact order/leverage details and get explicit confirmation before any trade, order cancellation/modification, or leverage change.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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
Findings (9)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
92% confidence
Finding

This endpoint enables direct cancellation of orders by order_id, but the module provides no explicit safety rule requiring ownership verification, order-detail display, or confirmation immediately before cancellation/modification. In a trading skill, weak parameter validation around destructive actions can let an agent cancel the wrong order if it uses stale context, guessed IDs, or attacker-supplied identifiers.

Content

Scanner excerpt · api-orders.md (reported line 98)May include surrounding context.


DELETE /orders/{order_id} — Cancel an order

Request

text

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · api-orders.md (reported line 102)May include surrounding context.

Request

text
DELETE /orders/ORD-20250325-001
X-API-KEY: {XCARD_TRADE_API_KEY}

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 29)May include surrounding context.

text

The key is stored in the environment variable `XCARD_TRADE_API_KEY`.
Base URL: `https://api.xcard.com/v2`

> Never expose, log, or repeat the API key in any response.
> If `XCARD_TRADE_API_KEY` is missing, ask the user to set it before proceeding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list includes very broad, common financial terms such as "balance," "margin," "wallet," and generic Chinese equivalents, which can cause the trading skill to activate in contexts where the user did not intend to access or discuss leveraged trading. In a crypto-perpetuals skill, unintended invocation is more dangerous because it can expose sensitive account data or steer the conversation toward high-risk trading actions without sufficiently clear user intent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The file documents endpoints that expose executed trades, closed positions, realized PnL, funding fees, and deposits/withdrawals, but it does not warn that these are sensitive financial records. Without clear notice, users may not understand the privacy implications of invoking the skill, which is especially risky in an agent that may proactively select tools based on trigger phrases.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases include very generic terms such as "history," "trade," and "pnl," plus common multilingual synonyms, making accidental loading of this module more likely during ordinary conversation. Because the file enables access to sensitive financial history, over-broad activation increases the risk that private trading or transaction data is fetched or surfaced without sufficiently specific user intent.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The module exposes /account/transactions with deposit and withdrawal history, which goes beyond the stated trading and margin-management scope and reveals sensitive account-funding activity. In an agent setting, this broadens data access to highly sensitive financial records and increases the chance of unnecessary disclosure when a user only intended trade-history assistance.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger list is broad enough to match ordinary finance or crypto discussion terms like 'price', 'quote', 'funding', and multilingual equivalents, which can cause the trading skill to activate when the user did not intend to invoke it. In a trading context, unintended activation is more dangerous because it can lead the agent into using account-scoped market/trading capabilities or presenting exchange-specific data in the wrong context, increasing the chance of erroneous follow-on actions.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list is very broad and includes common trading words like 'buy', 'sell', 'order', and 'stop', which can cause the skill to activate on ambiguous user requests. In a high-risk trading context, unintended invocation can route a casual discussion into order-execution flows and increase the chance of accidental destructive actions.

Content

No source excerpt is available for this finding.