ClawHub
Back to skill

Security audit

Ai video for free

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AI video workflow that uses third-party services and Telegram as advertised, but users should avoid sensitive content and review destinations before running it.

Install only if you are comfortable sending prompts, generated scripts, narration text, media, and the final video to HuggingFace, ElevenLabs, and Telegram. Use dedicated revocable API keys, confirm the Telegram chat ID before sending, avoid private or regulated content, and only use voice cloning with explicit permission from the speaker.

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 (6)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger phrases are broad enough to match ordinary requests to make a video, which can cause this skill to activate unexpectedly and start a workflow that sends content to third-party services. In this skill's context, overbroad activation is more dangerous because the pipeline performs external API calls and Telegram delivery automatically.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The description does not clearly warn that user prompts, generated scripts, audio text, and final media will be sent to HuggingFace, ElevenLabs, and Telegram. This creates a consent and data-handling risk because users may provide sensitive ideas or text without realizing it will leave the local system.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The example sends arbitrary text and an API key to a third-party TTS provider but gives no warning that user content leaves the local environment or may be processed and retained externally. In this skill context, users may submit sensitive prompts, scripts, or personal data for video generation, so omission of a privacy notice increases the risk of unintentional disclosure.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The guidance encourages voice cloning from Arabic recordings without any warning about consent, biometric sensitivity, impersonation risk, or legal/privacy obligations. Voiceprints are biometric data, and in a short-video automation skill this could enable unauthorized cloning of real people for generated media and distribution via Telegram.

Ssd 3

Medium
Confidence
96% confidence
Finding
The skill builds a Telegram caption from generated scene voiceovers, which may include user-provided sensitive content or unsafe model output, and transmits it to an external chat destination. This is dangerous because it expands the amount of text disclosed beyond the video itself and does so without minimization or a consent checkpoint.

External Transmission

Medium
Category
Data Exfiltration
Content
settings = emotion_map.get(emotion, emotion_map["محايد"])
    
    response = requests.post(
        f"https://api.elevenlabs.io/v1/text-to-speech/{voice_id}",
        headers={
            "xi-api-key": api_key,
Confidence
84% confidence
Finding
This code performs external transmission of user-provided text and includes an API key in the request headers to ElevenLabs. While expected for TTS functionality, it is still security-relevant because the skill automates content generation and may process sensitive user inputs without disclosure, classification, or safeguards.

Static analysis

No suspicious patterns detected.