Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed Alibaba DashScope media-generation helper; it sends user prompts and selected media to the Wan API but shows no hidden persistence, destructive behavior, or unrelated data access.
Install only if you are comfortable sending prompts, referenced URLs, and any explicitly provided local image files to Alibaba DashScope using your DASHSCOPE_API_KEY. Do not use private or confidential media unless your policy permits that upload.
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
print(f" Size : {args.size}")
print(f" Quantity: {args.quantity}")
resp = requests.post(url, headers=headers, json=payload, timeout=300)
result = resp.json()
if "code" in result:
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
print(f" Size : {args.size}")
print(f" Quantity: {args.quantity}")
resp = requests.post(url, headers=headers, json=payload, timeout=300)
result = resp.json()
if "code" in result:
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
print(f" Size : {args.size}")
print(f" Duration: {args.duration}s")
resp = requests.post(url, headers=headers, json=payload, timeout=60)
result = resp.json()
if "code" in result:
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
print(f" Size : {args.size}")
print(f" Duration: {args.duration}s")
resp = requests.post(url, headers=headers, json=payload, timeout=60)
result = resp.json()
if "code" in result:
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
print(f" Size : {args.size}")
print(f" Duration: {args.duration}s")
resp = requests.post(url, headers=headers, json=payload, timeout=60)
result = resp.json()
if "code" in result:
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
print(f"[text2video-get] Checking task: {args.task_id}")
resp = requests.get(url, headers=headers, timeout=60)
result = resp.json()
if "code" in result:
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
print(f"[text2video-get] Checking task: {args.task_id}")
resp = requests.get(url, headers=headers, timeout=60)
result = resp.json()
if "code" in result:
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
print(f"[text2video-get] Checking task: {args.task_id}")
resp = requests.get(url, headers=headers, timeout=60)
result = resp.json()
if "code" in result:
The skill invokes Python, requires an API key, and interacts with external model APIs, but it does not declare an explicit tool scope such as allowed tools or permissions. That weakens review and containment, because users and orchestrators cannot clearly see that the skill needs code execution, environment access, and network egress before running it.
The documentation states that local image files are converted to base64 and sent to the model API, but it does not clearly warn users that supplying a local file causes data to leave the machine and be uploaded to a third-party service. This can lead to accidental exfiltration of sensitive local content, especially because the feature accepts arbitrary local image paths.
The image2video feature allows a local image path and notes that the script will encode and pass it to the external API, but the skill text does not prominently disclose this as a privacy and data-transfer risk. Users may reasonably assume a local image is processed locally, resulting in unintended disclosure of private images to a remote provider.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
POST https://dashscope-intl.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generationPOST https://dashscope-us.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generationcurl --location 'https://dashscope.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
This markdown file includes a curl example that uploads user-supplied text and one or more image references to an external HTTP endpoint, but it does not warn users that their content is being sent to Alibaba Cloud for processing. For a skill description or API reference, omitting a clear disclosure about external transmission of potentially sensitive user data can leave users unaware of privacy implications.
This markdown file documents POST requests to external DashScope endpoints and instructs users to send prompts plus reference_urls, which may contain user media or personal data. While it explains how to call the API, it does not warn that prompts, media references, and authentication data are being transmitted to a third-party service.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
> 创建成功后使用返回的 `task_id` 查询结果，task_id 有效期24小时。请勿重复创建任务。
### curl 示例
#### 多角色互动（参考图像和视频）
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
POST https://dashscope-intl.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generationPOST https://dashscope-us.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generationcurl --location 'https://dashscope.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
创建成功后使用返回的
task_id查询结果，task_id 有效期24小时。请勿重复创建任务。
curl --location 'https://dashscope.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis' \
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
创建成功后使用返回的
task_id查询结果，task_id 有效期24小时。请勿重复创建任务。
curl --location 'https://dashscope.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis' \
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
print(f" Size : {args.size}")
print(f" Quantity: {args.quantity}")
resp = requests.post(url, headers=headers, json=payload, timeout=300)
result = resp.json()
if "code" in result:
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
print(f" Size : {args.size}")
print(f" Quantity: {args.quantity}")
resp = requests.post(url, headers=headers, json=payload, timeout=300)
result = resp.json()
if "code" in result:
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
print(f" Size : {args.size}")
print(f" Quantity: {args.quantity}")
resp = requests.post(url, headers=headers, json=payload, timeout=300)
result = resp.json()
if "code" in result:
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
print(f" Size : {args.size}")
print(f" Quantity: {args.quantity}")
resp = requests.post(url, headers=headers, json=payload, timeout=300)
result = resp.json()
if "code" in result:
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
print(f" Size : {args.size}")
print(f" Quantity: {args.quantity}")
resp = requests.post(url, headers=headers, json=payload, timeout=300)
result = resp.json()
if "code" in result:
The image-edit flow sends user prompts and referenced image content to a third-party API, but the script does not provide an explicit privacy or data-transmission warning before upload. In a skill context that may be run over local user files, this can lead to unintentional disclosure of sensitive image content or prompts to an external service.
The image2video path base64-encodes a local image and sends the full file contents off-host to the external API without an explicit warning or confirmation. This increases the chance that users unknowingly upload private local media, which is especially relevant for a media-processing skill that directly accepts filesystem paths.
No suspicious patterns detected.