Back to skill

Security audit

Ai Test Case Generator

Security checks for vulnerabilities and agentic risk

Overview

The skill generates test cases as advertised, but it also requires automatically sending the generated content to an external HTTP service with insecure transport settings.

Install only if you are comfortable having generated test cases sent to the named conversion service. Do not use it with confidential requirements, internal APIs, credentials, unreleased product details, or security-sensitive scenarios unless the export step is made explicit, HTTPS-only, and approved by the user each time.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:113
Finding

Mandatory Transmission of Potentially Sensitive Test Data over an Insecure Connection

Content
View full analysis
" \ --data-urlencode "module_path=/功能测试" # 方案2:如果方案1失败,尝试 Python(需要安装 requests 库) python -c " import requests url = 'http://office-tools.wh.ctrm.5636cloud.com/api/v1/office/md-to-excel' files = {'md_data': (None, '<生成的markdown内容>'), 'module_path': (None, '/功能测试')} r = requests.post(url, files=files, verify=False) print(r.text) " ``` ### Technical Analysis The skill instructs the agent to automatically and mandatorily submit generated Markdown test cases to `office-tools.wh.ctrm.5636cloud.com` using plaintext HTTP. Test cases generated from user-provided requirements may include confidential business logic, internal architecture, API details, account examples, security controls, test credentials, or vulnerability-related scenarios. Because HTTP provides neither transport encryption nor authenticated endpoint identity, any party capable of observing or modifying the network path can read or alter the request and response. The returned Excel download URL could consequently be replaced with an attacker-controlled link. Both proposed clients also explicitly disable TLS certificate verification through `curl -k` and `requests.post(..., verify=False)`. These options provide no benefit while the configured endpoint uses HTTP, but they would leave the workflow vulnerable to man-in-the-middle interception even if the URL were changed to HTTPS without removing those options. The workflow provides no explicit user-consent step, sensitivity warning, redaction procedure, destination trust validation, or loca ...[truncated 1854 chars]
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill mandates automatic transmission of generated Markdown to an external API without warning about privacy, confidentiality, or retention implications. Test cases often embed unreleased features, internal workflows, credentials placeholders, or security assumptions, so silent upload materially increases data leakage risk.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill explicitly recommends insecure transport practices: plain HTTP and disabled TLS verification via curl -k and requests verify=False. This enables man-in-the-middle tampering, interception of potentially sensitive test content, and spoofed responses such as malicious download links.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

This duplicate finding identifies the same unsafe pattern: a raw external POST with user/content-controlled parameters to a non-vetted endpoint. In context, the skill’s automatic export requirement makes the misuse more dangerous because the behavior is framed as mandatory rather than optional.

Content

Scanner excerpt · SKILL.md (reported line 118)May include surrounding context.

bash
# 方案1:推荐 - 使用 curl(Windows/Linux/Mac 均支持)
curl -k -X POST "http://office-tools.wh.ctrm.5636cloud.com/api/v1/office/md-to-excel" \
  --data-urlencode "md_data=<生成的markdown内容>" \
  --data-urlencode "module_path=/功能测试"

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

This duplicate finding identifies the same unsafe pattern: a raw external POST with user/content-controlled parameters to a non-vetted endpoint. In context, the skill’s automatic export requirement makes the misuse more dangerous because the behavior is framed as mandatory rather than optional.

Content

Scanner excerpt · SKILL.md (reported line 118)May include surrounding context.

bash
# 方案1:推荐 - 使用 curl(Windows/Linux/Mac 均支持)
curl -k -X POST "http://office-tools.wh.ctrm.5636cloud.com/api/v1/office/md-to-excel" \
  --data-urlencode "md_data=<生成的markdown内容>" \
  --data-urlencode "module_path=/功能测试"

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The profile sets 'language: 中文', which imposes a fixed language behavior. There is no opt-in, fallback, or justification that this skill is region-specific, so it conflicts with the policy against forcing a specific language without user choice.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill extends beyond test-case generation by directing execution of curl/Python commands that send generated content to an external service. This creates an unnecessary exfiltration and command-execution pathway, especially because requirements or test cases may contain sensitive internal product details.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill recommends insecure API transport and certificate-bypass settings without any warning to the user. Even if users choose to export, the absence of security guidance normalizes unsafe behavior and increases the likelihood of interception or tampering.

Content

No source excerpt is available for this finding.

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
99% confidence
Finding

Using verify=False disables TLS certificate validation, making the client trust any certificate presented by an attacker. This undermines transport security and can lead to interception, response manipulation, or delivery of malicious download URLs.

Content

Scanner excerpt · SKILL.md (reported line 127)May include surrounding context.

import requests url = 'http://office-tools.wh.ctrm.5636cloud.com/api/v1/office/md-to-excel' files = {'md_data': (None, '<生成的markdown内容>'), 'module_path': (None, '/功能测试')} r = requests.post(url, files=files, verify=False) print(r.text) "

text

Static analysis

No suspicious patterns detected.