Back to skill

Security audit

'提取高优先级测试用例'

Security checks for vulnerabilities and agentic risk

Overview

This skill performs disclosed local Excel test-case filtering and formatting changes, with no evidence of hidden network access, persistence, credential use, or destructive behavior beyond the intended output workbook transformation.

Install only if you are comfortable with it creating modified output workbooks: it removes project/product columns, splits merged cells, filters rows, and rewrites case IDs. Keep the original workbook backed up, choose a distinct output filename, and avoid running the batch helper unless the hardcoded tmp folder contains only files you intend to process.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (9)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

声明描述的是一个会对测试用例Excel进行实质性处理和转换的技能:筛选特定优先级、删除列、重新编号,并保留格式。实际代码却只是读取一个固定路径的Excel文件,访问名为“登录”的工作表,并打印第1行表头,没有进行任何筛选、删除、重编号或保存操作。其主要目的更像是检查表头结构,而不是按声明执行测试用例整理,因此属于明显的描述与行为不匹配。

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

声明描述的是一个对测试用例Excel进行实际加工处理的技能,核心功能包括筛选P0/P1、删除列、重新编号并保持格式。但提供的代码片段只是一个验证脚本:加载已生成的最终版Excel,检查“登录”sheet表头和总列数,并据此判断某列是否已删除。它的主要目的与声明不一致,属于结果校验而不是执行所声明的数据处理流程。因此应判定为描述与代码行为不匹配。

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown file documents behaviors that alter workbook structure and data layout, including deleting columns, splitting merged cells, and renumbering cases. Although these actions are core functionality, the README does not clearly warn users that the output file will be structurally modified and may not preserve formulas, merge semantics, or original identifiers.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
85% confidence
Finding

The skill advertises file-processing behavior and includes code-oriented implementation guidance, but it does not declare any explicit tool scope such as permissions or allowed-tools. In an agent environment, missing scope boundaries can allow broader-than-necessary file access, increasing the risk of unintended reads of sensitive local files when the skill is invoked on user workbooks.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The manifest says the skill filters test-case Excel files, extracts P0/P1 cases, removes project/product columns, renumbers rows, and preserves formatting. This file merely opens a specific workbook, reads the first row of one sheet, and prints header values, with no filtering, column deletion, renumbering, or output generation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

The README presents the skill name, instructions, and example prompts entirely in Chinese, which can imply a fixed language requirement for invocation and use. There is no statement that other languages are supported or that Chinese is merely an example, so this may violate a language-choice policy.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

This markdown file describes operations such as deleting columns, unmerging cells, renumbering rows, and saving a processed output file. Although L210 recommends backing up the original file, it does not explicitly warn users that the skill performs data-altering transformations that may affect workbook structure and content.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest context states the skill should '删除项目/产品列' in addition to filtering P0/P1 cases and renumbering, but the package description string in setup.py only says it filters P0/P1 test cases, renumbers, and preserves formatting. This is a semantic mismatch in the skill's claimed behavior documentation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The package description at L06 is written entirely in Chinese, which imposes a specific language in user-facing metadata. There is no indication that the skill is intentionally region-specific or that alternative language support is offered, so this appears to violate the language/locale policy criteria.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.