Back to skill

Security audit

Expense Categorization Optimizer

Security checks for vulnerabilities and agentic risk

Overview

This skill provides local, informational expense-categorization guidance without accessing accounts or making transactions, though its triggers are broader than its title suggests.

Reasonable to install for informational expense-category templates and local analysis of text you provide. Treat tax, pricing, compliance, and financial-health suggestions as general guidance only, and be aware the broad trigger terms may activate it for financial questions outside expense categorization.

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
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The requirement 'Clear documentation in English' imposes a specific language policy in natural language. Under the policy criteria, forcing a language without user opt-in or a documented justification is a reportable locale/language violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill description prominently uses Chinese as the default instructional language, and the examples mix English and Chinese without stating that users may choose their preferred language. This can violate a language/locale policy when a skill imposes a specific language without opt-in or explicit justification.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list substantially exceeds expense categorization and reaches into broad financial analysis topics such as tax optimization, pricing strategy, receivables aging, invoice compliance, and financial health review. This can cause the agent to invoke the skill for unrelated or higher-stakes financial requests, leading to misleading guidance outside the skill’s stated scope and weakening routing and safety boundaries.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The top-level documentation says this is a descriptive skill providing templates and frameworks, with no real execution beyond that. However, the implementation parses financial keywords, extracts amounts and dates, computes aggregates, and generates tailored expense-categorization recommendations, which is a stronger analytical behavior than the docstring describes.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger list includes very generic terms such as "expense," "categorization," and "optimizer," which are likely to match ordinary user conversations unrelated to this specific skill. This can cause unintended invocation or routing collisions, making the skill activate in contexts where the user did not intend financial rule optimization and potentially exposing downstream systems to prompt hijacking or incorrect task handling.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The front matter sets language: en, which implies an English-language constraint, but the document later includes a full Chinese usage scenario. This inconsistency can amount to an unclear language policy because the skill does not explicitly state whether multilingual interaction is supported or user-selectable.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The manifest sets "language": "en", which can indicate a fixed language policy rather than allowing user preference or locale choice. Under the policy, language constraints should either be optional for the user or clearly justified as region-specific.

Content

No source excerpt is available for this finding.