Back to skill

Security audit

Finance Omni Risk

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Chinese financial risk-management guidance skill with no executable install surface or hidden system access, though users should treat its risk and legal-adjacent recommendations as advisory only.

Install only if you need mainland China financial risk-management guidance. Do not paste real customer identifiers, account details, internal ratings, training data, or confidential reports into the assistant; have qualified legal, compliance, and risk committees approve any actions such as freezing credit, litigation preservation, debt restructuring, or bankruptcy-related steps.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file states that when users ask in other languages, the skill will still respond primarily in Chinese. This is a language-policy constraint presented as mandatory behavior rather than a user-selectable option, which matches the locale/language policy violation criteria.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill explicitly says it does not replace legal or disposition decision-making, but later provides concrete operational escalation steps including litigation preservation and bankruptcy filing. In a regulated financial-risk workflow, that contradiction can cause an agent or user to treat the skill as authorized to recommend or operationalize legal actions beyond its declared boundary, increasing the chance of improper escalation or unauthorized decision support.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.