ClawHub
Back to skill

Security audit

Jimeng Video

Security checks for vulnerabilities and agentic risk

Overview

This skill is a purpose-aligned Seedance/Volcengine video-generation helper, but users should understand that prompts and media may be sent to a remote API using their own API key.

Install this only if you are comfortable using Volcengine/Seedance for remote generation. Treat prompts, uploaded images, videos, and audio as data sent to a third-party provider, and avoid secrets, private personal data, or confidential business assets unless your organization approves that use.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill explicitly documents sending user prompts and generated content to a third-party API endpoint, but it does not prominently disclose the privacy and data-handling implications to users. In a content-generation skill, prompts may contain sensitive business, personal, or copyrighted material, so omission of a clear external-transmission warning can lead to unintended data exposure.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The trigger phrase at line 11 (e.g. a generic video-creation invocation) is broad enough that the skill may be selected for ordinary requests unrelated to this specific Jimeng/Volcengine capability. Because the skill requires high-privilege API credentials and can initiate external media-generation actions, overbroad activation increases the chance of unintended invocation, unnecessary credential use, and unexpected third-party API calls.

Vague Triggers

Low
Confidence
77% confidence
Finding
The manifest exposes several triggers but does not define scope constraints, disambiguation rules, or exclusions for when the skill should not run. In a multi-skill environment, this can cause ambiguous routing and accidental activation of a credentialed external-service skill for general video-related prompts, creating unnecessary operational and privacy risk.

External Transmission

Medium
Category
Data Exfiltration
Content
### 生成带声音的视频

```bash
curl -X POST "https://ark.cn-beijing.volces.com/api/v3/contents/generations/tasks" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${API_KEY}" \
  -d '{
Confidence
87% confidence
Finding
The curl example shows direct transmission of user-supplied content to an external service, which is expected for this skill's functionality but still creates a real data-exposure boundary. Because the skill is specifically for remote AI video generation, this is contextually legitimate, yet dangerous if users assume processing is local or do not realize prompts and assets leave their environment.

Static analysis

No suspicious patterns detected.