Back to skill

Security audit

木工厂报价助手

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly matches a woodworking quotation workflow, but its Excel export silently embeds fixed third-party branding/contact details and accepts spreadsheet content in a way that can create unsafe formulas.

Review this skill before installing. It is not showing OS-level persistence or exfiltration, but you should remove or configure the hardcoded business name/contact, sanitize Excel cell values, and avoid exporting customer-facing quotations until those issues are fixed.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T01 · Skill Instruction Hijacking

Warning
Location
scripts/export_quotation.py:60
Finding

Undeclared Business Branding and Contact Information Injected into Exported Quotations

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/export_quotation.py:45
Finding

Excel Formula Injection Through Untrusted Quotation Fields

Content
View full analysis
<输出路径.xlsx>") print("示例: python export_quotation.py '{\"status\":\"success\",...}' ./output.xlsx") sys.exit(1) quotation = json.loads(sys.argv[1]) output_path = sys.argv[2] if quotation.get("status") == "error": print(f"❌ 导出失败:{quotation.get('message')}") sys.exit(1) path = create_quotation_workbook(quotation, output_path) print(f"✅ 报价单已导出:{path}") ``` ### Technical Analysis The exporter treats quotatio ...[truncated 3130 chars]
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (17)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

声明描述的核心能力是“报价助手”,重点在成本核算与报价单生成;图纸尺寸提取只是其中一个支持能力。而实际代码几乎全部围绕图纸文件解析:PDF 文本提取、尺寸正则识别、材料/工艺关键词提取,以及对图片描述的简单解析。没有任何价格表、损耗率、成本公式、报价逻辑、表格生成或导出实现。虽然代码与“支持图纸尺寸提取”这一子功能部分一致,但其主要行为与声明的主要用途存在明显偏差,因此应判定为描述与行为不匹配。

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The trigger phrases are broad and common enough that ordinary user requests could unintentionally activate the skill. In an agent environment, overbroad activation can cause the skill to process files, drawings, or pricing data in contexts where the user did not explicitly intend to invoke it, increasing the chance of unintended data handling or workflow execution.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The keyword and example sections define activation by loose natural-language phrases without specifying scope, exclusions, or confirmation requirements. This makes accidental or adversarial prompt overlap more likely, which can trigger parsing or quotation workflows on sensitive business documents without sufficiently clear user intent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The skill documentation is entirely written as Chinese-only operational guidance, which indicates a language-specific constraint without any stated user choice, opt-in, or justification for enforcing that locale. Under the policy, forced language behavior should be documented as optional or clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This code file contains natural-language descriptions and output strings exclusively in Chinese, including the module description and later user-facing error/report text. Under the policy, forcing a specific language without opt-in is a natural-language policy violation unless the locale restriction is explicitly documented and justified, which it is not here.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

L035注释声明当前分支是“公开版:示例配置”,其材料价格键仅包含“板材(18mm)”等公开版名称;但在L195-L196中,无材料输入时默认使用“桦木板材(18mm)”,该名称只存在于内部版配置。结果是在EXTERNAL模式下默认路径会生成空材料明细和0材料成本,和代码注释所表达的公开版报价逻辑相矛盾。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The module description and all user-facing strings are written only in Chinese, which imposes a specific language on users. Under the policy, forcing a language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly justified, which this file does not do.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This code file contains natural-language docstrings, comments, and CLI usage/output entirely in Chinese, including the primary user-facing invocation message. Under the policy, forcing a specific language without user opt-in is a locale/language policy violation unless the constraint is explicitly justified, which is not present here.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The module docstring and parse_dwg docstring present DWG parsing as a supported capability, yet the function explicitly returns a stub response with status 'pending' and a note asking for a future implementation. This is an active contradiction between the code documentation and actual behavior.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The docstring and module header describe support for parsing image drawings and extracting dimensions/materials/processes, but parse_image never opens or analyzes the image file at all. Instead, it derives all results solely from the description string, which directly contradicts the stated behavior of image-based parsing.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The listed core trigger phrases such as “算报价”, “图纸报价”, and “生成报价表” are short, generic phrases that could easily appear in normal workplace chat, while the guide does not define exclusion conditions or a narrow activation context. This creates ambiguity about when the skill should activate versus when the user is merely discussing pricing work.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The guide explicitly encourages uploading drawing files and exporting/downloading archived quotation spreadsheets, but it does not warn users about how those files are stored, where exports are written, or how sensitive design and pricing data is handled. In a business context, drawings, procurement prices, and quotations can contain confidential commercial information, so silent file creation or retention increases the risk of unintended disclosure or mishandling.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill advertises one-click Excel export but does not clearly warn that quotation details will be written to a file. In this business context, exported files may contain sensitive pricing, cost structure, or supplier-related data, so silent file creation can lead to unintended persistence or disclosure.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

This markdown file contains only Chinese-language instructions and does not indicate that the user can choose another language or that the skill is intentionally limited to a Chinese-speaking audience. Under the policy rule for language or locale constraints, forcing a specific language without opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The file presents all user-facing instructions and labels exclusively in Chinese, with no indication that the skill is region-specific or that users may opt into another language. Under the policy rule for natural-language violations, forcing a specific language without user choice can be a locale-policy issue.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

This markdown file uses Chinese exclusively for headings, table labels, and usage instructions. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale constraint is explicitly documented and justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

All user-facing guidance, labels, and trigger examples are written exclusively in Chinese, and the file does not state that the skill is intentionally limited to Chinese-speaking users or a China-specific deployment. Under the policy, forcing a specific language without opt-in or justification is a natural-language policy concern.

Content

No source excerpt is available for this finding.