Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward SiliconFlow image-generation guide that uses a user-provided API key and sends image prompts to the expected third-party service.
Install only if you are comfortable using SiliconFlow as the image provider. Treat prompts as data sent to a third party, keep your API key private, and avoid including secrets, personal identifiers, confidential work, or regulated data in image prompts.
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.
emoji: "🎨"
requires:
bins:
- curl
requires_env:
- name: SILICONFLOW_API_KEY
description: 硅基流动 API Key，在 cloud.siliconflow.cn 注册后创建，新用户有免费额度
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.
emoji: "🎨"
requires:
bins:
- curl
requires_env:
- name: SILICONFLOW_API_KEY
description: 硅基流动 API Key,在 cloud.siliconflow.cn 注册后创建,新用户有免费额度
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
---
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
emoji: "🎨"
requires:
bins:
- curl
requires_env:
- name: SILICONFLOW_API_KEY
description: 硅基流动 API Key，在 cloud.siliconflow.cn 注册后创建，新用户有免费额度
The activation examples include generic phrases like “画一张...” and “生成一张[风格][内容]的图”, which are common conversational requests and may match ordinary chat rather than deliberate skill invocation. The section does not provide exclusion conditions or negative examples to clarify when this skill should not activate.
The skill sends user-supplied prompts to a third-party API but does not clearly disclose that prompt content leaves the local environment and is processed by SiliconFlow. If users include sensitive personal, business, or regulated information in prompts, that data could be exposed to an external service without informed consent.
This command transmits the user's prompt to an external image-generation endpoint. While this is core functionality rather than malicious behavior, it becomes a real security/privacy issue if users are not warned that their content is being shared with a third party and may contain sensitive information.
curl -s -X POST "https://api.siliconflow.cn/v1/images/generations" \
-H "Authorization: Bearer $SILICONFLOW_API_KEY" \
-H "Content-Type: application/json" \
-d '{
The Kolors example also sends prompt content, including potentially revealing descriptive text, to SiliconFlow's API. In context this is intended behavior, but it still represents external disclosure risk if the skill is used with confidential or personal data.
curl -s -X POST "https://api.siliconflow.cn/v1/images/generations" \
-H "Authorization: Bearer $SILICONFLOW_API_KEY" \
-H "Content-Type: application/json" \
-d '{
This is another explicit third-party transmission of user content to a remote image-generation service. The risk is contextual rather than inherently malicious: the skill normalizes sending arbitrary prompts off-box without sufficient warning or consent language.
curl -s -X POST "https://api.siliconflow.cn/v1/images/generations" \
-H "Authorization: Bearer $SILICONFLOW_API_KEY" \
-H "Content-Type: application/json" \
-d '{
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
所有模型均使用同一接口：
POST https://api.siliconflow.cn/v1/images/generations
This markdown file presents all instructions and model guidance only in Chinese, which can constitute a language/locale policy issue when no user opt-in or alternative language option is provided. The content does not indicate that the skill is region-specific or that Chinese-only output is an intentional, justified constraint.
No suspicious patterns detected.