Back to skill

Security audit

Steering Gear

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed steering-gear drawing workflow that uses jixietools.com APIs and creates a guest production sheet, with no hidden local code or credential access found.

Before installing, understand that the skill sends steering-gear design inputs to jixietools.com, creates a guest production sheet there, and polls that service for drawing status. Avoid using it for confidential engineering designs unless you are comfortable sharing those parameters with that service.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (11)

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 29)May include surrounding context.

Step 1: 列出转向器产品

  1. 用 curl 获取转向器列表:
    bash
    curl -s "https://jixietools.com/api/v1/products?category_id=8" | python3 -m json.tool
    

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 29)May include surrounding context.

Step 1: 列出转向器产品

  1. 用 curl 获取转向器列表:
    bash
    curl -s "https://jixietools.com/api/v1/products?category_id=8" | python3 -m json.tool
    

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 47)May include surrounding context.

用户选择后,获取该产品的参数定义:

bash
curl -s "https://jixietools.com/api/v1/products/PRODUCT_ID/start" | python3 -m json.tool

返回结构:

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Although this is not literal script fetching, the skill operationalizes outbound POST requests that mutate state on an external service using persistent identifiers. In the skill context, this is more dangerous than a read-only fetch because it can change remote records and continue an unauthenticated workflow outside the user's immediate awareness.

Content

Scanner excerpt · SKILL.md (reported line 147)May include surrounding context.

  1. 收集要修改的参数名和新值
  2. 构建增量请求:只包含变化的参数 + filename:
    bash
    curl -s -X POST "https://jixietools.com/api/v1/products/PRODUCT_ID/calculate" \
      -H "Content-Type: application/json" \
      -d '{"inputs": {"修改的参数名": "新值"}, "filename": "之前保存的filename"}'
    

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The description says to use the skill when the user says "转向器", "生成转向器图纸", "做一个转向器", or "steering-gear". Phrases like "转向器" and especially "做一个转向器" are broad enough to match ordinary conversation without clearly constraining the context, and the file provides no negative examples or exclusion conditions.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill does not clearly warn users that their design parameters will be transmitted to a third-party service and may later be used to create a guest-access production sheet. Lack of upfront notice undermines informed consent and can expose potentially sensitive engineering data or workflow metadata to external systems unexpectedly.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
88% confidence
Finding

This step sends user-related parameter data to an external calculation endpoint before the user has been clearly warned about third-party transmission. Even though the payload may seem limited, it establishes a pattern of exporting design-session data to an external service without explicit informed consent.

Content

Scanner excerpt · SKILL.md (reported line 78)May include surrounding context.

  • 有 options_source 的参数(下拉选项类型): 先进行一次预计算获取 dropdown 选项:
    bash
    curl -s -X POST "https://jixietools.com/api/v1/products/PRODUCT_ID/calculate" \
      -H "Content-Type: application/json" \
      -d '{"inputs": {"有选项的参数名": ""}}'
    

External Transmission

Medium
Category
Data Exfiltration
Confidence
93% confidence
Finding

The incremental update call transmits modified parameters and a persistent filename token to an external service, linking multiple user interactions into a durable external record. This increases privacy and integrity risks because design revisions and session identifiers are continuously exported without strong consent boundaries or authentication guarantees.

Content

Scanner excerpt · SKILL.md (reported line 147)May include surrounding context.

  1. 收集要修改的参数名和新值
  2. 构建增量请求:只包含变化的参数 + filename:
    bash
    curl -s -X POST "https://jixietools.com/api/v1/products/PRODUCT_ID/calculate" \
      -H "Content-Type: application/json" \
      -d '{"inputs": {"修改的参数名": "新值"}, "filename": "之前保存的filename"}'
    

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill is presented as a CAD drawing generator, but it also initiates and monitors external production orders, expanding its capabilities beyond what users are likely to expect. This creates a transparency and consent problem because user-provided engineering parameters are not only used for calculation but also used to trigger unauthenticated order workflows on a third-party service.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill exposes unauthenticated guest creation and status tracking for production orders, which can be abused to create external records without strong user identity or authorization checks. In a chat-agent context, this is more dangerous because the agent can operationalize external actions at scale, potentially causing spam, unauthorized submissions, or leakage through guessable/shared guest links.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The instruction "用中文与用户对话" mandates a specific language for all interactions. Under the policy, forcing a language or locale without opt-in or clear documented justification is a natural-language policy violation.

Content

No source excerpt is available for this finding.