ClawHub
Back to skill

Security audit

Nuwa Video Gen - MiniMax女娲视频

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward MiniMax video-generation helper, but users should understand that prompts and image or face-reference URLs are sent to MiniMax.

Install only if you trust this MiniMax API workflow. Use a dedicated API key where possible, monitor billing or credits, avoid confidential prompts and private image URLs, and only use face-reference images with the subject's permission.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Tainted flow: 'download_url' from requests.get (line 124, network input) → requests.get (network output)

Medium
Category
Data Flow
Content
print(f"正在下载视频到 {output_path}...")
    with open(output_path, "wb") as f:
        video_response = requests.get(download_url)
        video_response.raise_for_status()
        f.write(video_response.content)
    print(f"视频已成功保存至 {output_path}")
Confidence
92% confidence
Finding
The code fetches a download URL obtained from a remote API response and then performs a second request to that URL without validating the scheme, host, or destination. If the upstream response is malicious, compromised, or manipulated, this can enable server-side request forgery behavior or downloading attacker-controlled content, which is then written directly to disk.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs users to send prompts, image URLs, and in one mode face images to a third-party API, but it does not clearly disclose that this data leaves the local environment and may be processed, stored, or logged by an external provider. This is more concerning here because the skill explicitly supports subject-reference generation using human face photos, which raises privacy and consent risks beyond ordinary prompt transmission.

Static analysis

No suspicious patterns detected.