Security audit
Security checks for vulnerabilities and agentic risk
This skill is not clearly malicious, but it needs review because it generates detailed electrical and plumbing construction documents without strong professional-safety limits.
Review this before installing if you might use it for real renovation work. Treat generated construction, electrical, plumbing, material, and code-compliance outputs as conceptual drafts only, have licensed local professionals verify them, and avoid sending sensitive floor plans, client details, or private room data to remote image services unless you accept those providers' terms. Review any optional sudo or third-party setup commands before running them.
The skill advertises generation of complete construction documents, including electrical and plumbing routing, as if they are ready-to-use outputs, but it does not clearly warn that these are not a substitute for review by licensed professionals. In a home-renovation context, users may rely on inaccurate or noncompliant wiring/plumbing plans, creating material safety and legal risks.
Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.
--style "欧式豪华"
--living-room "22.5㎡"
--bedroom "20㎡"
--output prompts.md
### 生成施工图纸套装 🆕
The guide encourages use of third-party APIs and online image-generation services, including API key configuration, but does not warn users that prompts, room details, and other potentially sensitive design data will be transmitted to external providers. In a home-design context, prompts may include floor plans, room sizes, layouts, or client information, creating a real privacy and data-handling risk if users assume processing is local or confidential.
The guide directs users to register or log in to third-party AI services using personal accounts such as WeChat or QQ, but does not warn that prompts, images, account data, and potentially project-related content may be uploaded to external providers. In a skill context, this can cause unintended disclosure of user data or sensitive design materials to services with separate privacy policies and retention practices.
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
# 1. 安装依赖
sudo apt update
sudo apt install -y git python3 python3-pip python3-venv
# 2. 克隆 WebUI
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
# 1. 安装依赖
sudo apt update
sudo apt install -y git python3 python3-pip python3-venv
# 2. 克隆 WebUI
The trigger list is very broad and includes generic home-design phrases without clear gating conditions, which can cause the skill to activate for loosely related requests. In an agent setting, overly broad activation increases the chance of unintended invocation, confusing handoffs, or the skill being used in contexts where its assumptions and safeguards do not apply.
The description is primarily written in Chinese and frames the skill around Chinese construction standards and terminology by default, without offering the user a language or locale choice. This can violate language/locale policy where users should be allowed to choose their preferred language or region unless the constraint is clearly documented as region-specific.
The trigger phrases are broad enough to activate on generic design or layout requests without clearly limiting the skill to conceptual interior-design assistance. In context, that increases the chance the agent will invoke a skill that can generate detailed construction and wiring guidance when the user did not explicitly request regulated or safety-sensitive output.
The module title and description are written entirely in Chinese, and the CLI help text throughout the file is also Chinese-only. This effectively forces a specific language/locale for users without any opt-in, which matches the language-policy violation criteria.
Suspicious Unicode normalization or mixed-script content
This Python file contains natural-language content entirely in Chinese and presents a fixed Chinese-speaking persona/output style in generated artifacts. Under the policy rule for language/locale constraints, forcing a specific language without user opt-in is a natural-language policy violation.
The generated electrical plan content is fully hardcoded in Chinese, with no mechanism for user selection of language or locale. This creates a mandatory language constraint rather than an opt-in or documented region-specific behavior.
The summary output is entirely in Chinese and explicitly identifies the designer as 'AI 助理 - 小美', reinforcing a fixed-language user experience with no opt-in. This violates the language/locale policy because users are not offered any language choice or documented justification for the restriction.
The file's title, description, CLI help text, generated document headings, and output messaging are all hard-coded in Chinese, indicating the skill operates in a single language. There is no user opt-in, language selection parameter, or documentation that this is intentionally limited to a Chinese-language or region-specific use case.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
}
try:
response = requests.post(f"{sd_url}/sdapi/v1/txt2img", json=payload, timeout=120)
response.raise_for_status()
result = response.json()
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
}
try:
response = requests.post(
"https://api.openai.com/v1/images/generations",
headers=headers,
json=payload,
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
}
try:
response = requests.post(
"https://api.openai.com/v1/images/generations",
headers=headers,
json=payload,
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
}
try:
response = requests.post(
"https://api.openai.com/v1/images/generations",
headers=headers,
json=payload,
This code posts the user's prompt to the OpenAI image generation API, which transmits user-supplied content to an external service. Although the function docstring names the backend, there is no visible confirmation prompt or user-facing disclosure in the CLI flow warning that prompts may leave the local system.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
try:
response = requests.post(
"https://api.openai.com/v1/images/generations",
headers=headers,
json=payload,
timeout=120
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
}
try:
response = requests.post(
"https://api.liblib.ai/v1/generate",
headers=headers,
json=payload,
The function sends prompt and negative prompt data to a third-party LiblibAI endpoint. The script does not display any warning or confirmation to users that their generation input will be shared with an external provider when this backend is selected.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
try:
response = requests.post(
"https://api.liblib.ai/v1/generate",
headers=headers,
json=payload,
timeout=120
Suspicious Unicode normalization or mixed-script content
No suspicious patterns detected.