ClawHub
Back to skill

Security audit

Security App Wealth Advisor

Security checks for vulnerabilities and agentic risk

Overview

This is a document-only wealth-advisory design aid with clear repeated limits against real customer recommendations, orders, persistence, or data access, though one summary block is broader than the rest of the skill.

Before installing, treat this as an internal drafting and review reference only. Do not paste real customer financial records into the conversation, do not use outputs as direct personalized advice, and have qualified compliance or advisory staff review any scripts, matching rules, or workflow designs before use. The publisher should tighten the short English/Chinese summary so it matches the narrower non-customer-facing scope used elsewhere.

A.I.G

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

SkillSpector

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)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The English summary says the skill provides 'product recommendations' and 'investment consultation', which conflicts with the rest of the document that repeatedly restricts the skill to design-time/review-time assistance and forbids direct personalised advice. In a regulated financial context, this mismatch can cause users or downstream agents to rely on the broader English description and use the skill for customer-facing recommendations, creating suitability and compliance risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
77% confidence
Finding

The English and Chinese summary text presents the skill as built for bank mobile applications and wealth-management advisors in China, and describes it as providing product recommendations and investment consultation. This is a natural-language locale targeting constraint without any user opt-in or explicit justification as a region-specific compliance limitation in that specific summary block.

Content

No source excerpt is available for this finding.