ClawHub
Back to skill

Security audit

财务分析师

Security checks for vulnerabilities and agentic risk

Overview

This is a finance-analysis persona skill with broad activation terms, but no evidence of hidden code, exfiltration, destructive actions, or privilege abuse.

Install this skill if you want a Chinese-language finance analysis assistant. Be mindful that its broad triggers may activate during general report, budget, or analysis discussions, and avoid sharing sensitive business data unless you intend to use the finance-analysis workflow.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (6)

Vague Triggers

High
Confidence
98% confidence
Finding
The trigger set is broad and consists of common business terms, so the skill may activate during ordinary conversations unrelated to invoking a finance specialist. This can cause unintended routing, expose user business data to the wrong specialist context, and degrade trust by producing unsolicited finance-oriented behavior.

Natural-Language Policy Violations

Medium
Confidence
87% confidence
Finding
The skill name, description, and the full operating instructions are written exclusively in Chinese and specifically oriented to serving Chinese enterprises, but the file does not state that language is optional or user-selected. This can constitute a language/locale policy issue because the skill appears to force a specific language/locale experience without opt-in.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
91% confidence
Finding
The trigger '财务' is extremely short and high-frequency, making accidental activation likely whenever users discuss finance-related topics in general. In this skill's context, that increases the chance of unintended specialist routing and unnecessary handling of potentially sensitive financial discussions.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
94% confidence
Finding
The trigger '分析' is especially broad because it applies to many unrelated domains, making false activation highly probable. This is dangerous operationally because it can hijack ordinary analytical conversations and route them into a finance-specific persona without user intent.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
89% confidence
Finding
The trigger '报表' is generic and may match routine discussion of reports in HR, sales, operations, or engineering contexts. This can cause unintended invocation of the finance skill and mis-handle context that should remain with another workflow or assistant role.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
87% confidence
Finding
The trigger '预算' is common across many non-finance tasks, including project planning and personal budgeting, so it can activate unintentionally. In a skill designed for enterprise financial analysis, this creates avoidable confusion and may pull unrelated sensitive planning discussions into the wrong specialist flow.

Static analysis

No suspicious patterns detected.