Back to skill

Security audit

Market Inquiry Agent

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed sales-assistance workflow for parsing trade inquiries and drafting quotations, with no executable code or hidden install behavior found.

Before installing, be aware that this skill may process customer messages, prices, trade terms, certifications, and payment or bank details. Use it to draft and organize quotations, but manually verify recipient, language, pricing, legal or certification claims, payment terms, and attachments before sending anything externally.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Tp2

High
Category
MCP Tool Poisoning
Confidence
85% confidence
Finding

Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The language handling table routes several non-Chinese inquiries into Chinese-summary outputs by default, without clear user opt-in or customer-facing language control. In practice this can cause unintended exposure of inquiry contents to operators or downstream systems in Chinese, create privacy/compliance concerns for cross-border customer data handling, and increase the chance of sending the wrong-language or mistranslated business content.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The workflow explicitly allows users to '下载或直接发送' generated quotations, but the skill does not require a final review for sensitive customer data, pricing, bank details, Incoterms, or regulatory claims before transmission. In a sales/export context, automatic sending can leak confidential business information, send inaccurate terms to customers, or disclose personal data across jurisdictions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

This markdown reference is primarily written in Chinese, but the quotation template hard-codes English business text such as "QUOTATION," "Dear [客户]," and the terms section. Because the file does not state that English is optional or selected by the user, it can steer outputs toward a specific language without user opt-in.

Content

No source excerpt is available for this finding.