计件工资与工序单价核对（免费版）

• Outcome: Pass
• Latest audit: Sep 20, 2026 · 1:23 PM UTC
• Version: 1.0.49

Overview

This is a local payroll-table checker with clear limitations and no evidence of network use, persistence, credential access, or hidden execution.

Before relying on this for payroll, confirm the output’s checks_not_run list: the free version does not validate standard process rates, recompute whether minimum-wage top-up is correct, detect duplicate rework/scrap deductions, or provide legal/payroll compliance advice. It appears safe to run locally on wage-table text, but treat it as an arithmetic checker, not a full payroll approval control.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (8)

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence
• Finding: The skill’s metadata and user-facing description repeatedly imply it can detect issues such as incorrect process rates, duplicate rework deductions, and top-up errors, while later text admits those checks are not implemented in the free version. In a payroll context, that mismatch can cause operators to rely on the tool as a pre-pay control and miss material wage errors, leading to underpayment/overpayment, disputes, and false assurance.

Ae1

High

• Category: analysis-evasion

Content

node scripts/run.mjs --sample # 先看样例

• Confidence: 100% confidence
• Finding: Referenced artifact was not completely inspected

Ae1

High

• Category: analysis-evasion

Content

node scripts/run.mjs --sample # 先看样例

• Confidence: 100% confidence
• Finding: Referenced artifact was not completely inspected

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: An overly broad trigger phrase can cause the skill to activate in situations outside its narrow intended use, especially because the skill handles payroll-adjacent validation terms that may appear in general accounting conversations. That increases the chance of accidental invocation and misplaced trust in a specialized checker with known scope limitations.

Natural-Language Policy Violations

Medium

• Confidence: 95% confidence
• Finding: The natural-language comments and user-facing strings throughout the file are entirely in Chinese and describe the skill's interface, advice, and outputs in Chinese only. Under the stated policy, forcing a specific language without user opt-in is a locale/language policy violation unless clearly justified as region-specific, and this file does not explicitly document such a justification or offer a language choice.

Natural-Language Policy Violations

Medium

• Confidence: 96% confidence
• Finding: The file’s docstring, usage text, errors, and operational messages are entirely in Chinese, which imposes a specific language on users. The policy allows locale constraints only when clearly justified or when users are given a choice, neither of which is present in this file.

Intent-Code Divergence

Low

• Confidence: 87% confidence
• Finding: The inline documentation says non-JSON input may be '例如直接粘贴的合同全文', but the skill's stated purpose and the rest of the file consistently require a piece-rate wage table with headers. The code itself simply passes plain text through to the engine, so this comment broadens the apparent intended input type in a way that contradicts the skill's declared scope.

Natural-Language Policy Violations

Low

• Confidence: 92% confidence
• Finding: The file contains only Chinese-language tabular content and provides no indication that the locale is optional or intentionally limited to Chinese-speaking users. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation.

Static analysis

No suspicious patterns detected.