Back to skill

Security audit

Salary Tax Calculator

Security checks for vulnerabilities and agentic risk

Overview

The skill is a China personal income tax helper with expected local calculation/report behavior, but its documentation overstates some supported tax scenarios and users should verify results before relying on them.

Install only if you need a Chinese-language China individual income tax estimator. Treat outputs as estimates, verify current tax-year rules with official tax guidance or a professional, and avoid relying on the labor-income, annual-settlement, reverse salary, or other-income handling until those gaps are fixed.

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 Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (10)

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The skill description promises narrowly scoped tax-calculation behavior, but the analyzed behavior includes undisclosed local file reads/writes and feature gaps that do not match the stated functionality. This mismatch is dangerous because reviewers and users may trust the declared purpose while the skill performs additional side effects, which can conceal data exposure or unauthorized artifact generation.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
80% confidence
Finding

The skill instructs the agent to read local reference files and invoke a local script, but it does not declare any explicit tool scope or permissions boundary. That creates hidden file access expectations and weakens least-privilege controls, increasing the chance that broader file read/write capabilities are available than users or reviewers would expect.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The activation text is very broad and can trigger on ordinary tax-related conversation without a clear invocation boundary. Over-broad triggering can cause the skill to collect sensitive financial data or invoke scripts in contexts where the user did not intend to use this skill, increasing privacy and misuse risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This reference manual gives detailed tax calculation formulas, thresholds, and optimization advice that users may rely on as authoritative, but it does not clearly state that tax treatment depends on tax year, jurisdictional updates, policy changes, and individual circumstances. In a tax-focused skill, omission of such caveats increases the chance of users making incorrect filing or withholding decisions based on stale or incomplete guidance.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill advertises support for annual settlement estimation and labor-remuneration tax scenarios, but the implementation only covers salary withholding and year-end bonus comparison. In a tax-calculation skill, this mismatch can mislead users into relying on outputs for unsupported tax cases, producing incorrect tax advice and potentially causing underpayment, overpayment, or filing errors.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file's docstring, argument descriptions, report content, and console output are all hard-coded in Chinese, and the script provides no option to select another language or locale. This is a natural-language policy concern because it imposes a specific language on all users without opt-in or documented regional scope.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The program accepts and displays other_income but never incorporates it into any tax computation, creating a misleading report that appears comprehensive while omitting taxable income. In a financial/tax context this is especially dangerous because users may trust the report for compliance decisions, leading to materially inaccurate tax estimates.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
98% confidence
Finding

All user-facing instructions, examples, and descriptions are written exclusively in Chinese, and the skill does not state that users may interact in another language or choose their preferred locale. Under the policy, a skill should not impose a specific language without opt-in unless the constraint is clearly documented and justified as region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

A file that unconditionally presents content in a single language can violate language/locale policy when no choice or opt-in is offered. Here, the document forces Chinese-language output without any note that the skill is China-specific or that language is configurable.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

The top-level documentation presents a JSON configuration format as the expected input schema, but it does not list other_income even though the code accepts and uses that key for report content. This is not just incomplete implementation detail; it misstates the supported documented interface for inputs.

Content

No source excerpt is available for this finding.