Security audit
Security checks for vulnerabilities and agentic risk
This is a transparent design-to-code guide with expected local project scaffolding behavior, but users should review full-stack and third-party design-service use before running it.
Install only if you want an agent to build runnable local app projects from designs. Before running generated full-stack setup, confirm whether you want databases, admin panels, auth, payment, AI, file upload, or external Figma/API provider access, and review generated .env values and seed accounts before deployment.
The skill description is written entirely in Chinese and specifies the behavior and outputs in that locale, but nowhere offers a language choice or states that the skill is intentionally limited to Chinese users or a Chinese compliance context. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
- `POST /api/resources` — 创建
- `PUT /api/resources/:id` — 全量更新
- `PATCH /api/resources/:id` — 部分更新
- `DELETE /api/resources/:id` — 删除（或软删除）
- **认证与授权**：
- 用户注册/登录/登出 API
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
- `POST /api/resources` — 创建
- `PUT /api/resources/:id` — 全量更新
- `PATCH /api/resources/:id` — 部分更新
- `DELETE /api/resources/:id` — 删除（或软删除）
- **认证与授权**：
- 用户注册/登录/登出 API
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
npm install
# 2. 配置环境变量
cp .env.example .env
# 编辑 .env 填入数据库路径、JWT Secret 等
# 3. 初始化数据库（以 Prisma + SQLite 为例）
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
npm install
# 2. 配置环境变量
cp .env.example .env
# 编辑 .env 填入数据库路径、JWT Secret 等
# 3. 初始化数据库（以 Prisma + SQLite 为例）
The activation section says the skill triggers whenever a user asks to turn a design into code, then includes broad phrases like “帮我实现这个页面” and “根据这张图生成可运行的项目.” These phrases can easily occur in ordinary conversation and the file does not provide exclusion conditions or narrow context boundaries, so the skill could be invoked more broadly than intended.
The skill directs use of external services such as Figma MCP and model APIs for parsing design/content inputs, but provides no privacy notice or consent mechanism for sending user data to third parties. Design files, screenshots, and specs may contain proprietary or sensitive information, so silent transmission materially increases confidentiality risk.
The skill materially expands from design-to-code into full backend, database, auth, payment, and AI system generation. That scope creep increases the chance that an agent will generate sensitive capabilities the user did not explicitly request, including state-changing services and secret-handling code, which raises attack surface and the risk of unsafe default implementations.
The skill instructs creation, migration, seeding, and use of real databases and persistent storage without any user-facing warning that these actions can modify local state and create durable data. In agent-driven environments, this can cause unintended data creation, schema changes, and persistence side effects beyond what a user expects from a design-to-code workflow.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
Automatically deriving database schemas, CRUD APIs, and admin systems from UI nouns and flows causes the skill to invent application behavior beyond the provided design artifact. This can lead to over-privileged scaffolding, unnecessary data collection, and unsafe assumptions about entities, roles, and destructive operations that were never explicitly authorized by the user.
No suspicious patterns detected.