ClawHub
Back to skill

Security audit

video-content-watcher

Security checks for vulnerabilities and agentic risk

Overview

This video-analysis skill has a clear purpose, but it relies on unbundled local code while handling API keys and potentially private video transcripts.

Review before installing. Only use this skill if you can verify the separate video-reader project it imports and you are comfortable providing MiniMax and possibly Whisper/OpenAI credentials. Avoid private meetings, proprietary recordings, regulated content, or sensitive URLs unless you accept that derived content may leave your machine.

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 (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger description is broad enough to match many ordinary requests about videos, which can cause the skill to activate unexpectedly and process user-supplied video content without sufficiently explicit intent. In this skill, accidental activation matters because the workflow can download media, transcribe audio, and send transcript content to external services, increasing privacy and data-handling risk.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The documentation describes transcription and LLM analysis but does not clearly warn that video-derived content may be transmitted to third-party APIs such as MiniMax and possibly Whisper API. This omission is dangerous because users may provide private, copyrighted, or sensitive videos assuming local-only processing, leading to unintended external disclosure.

External Transmission

Medium
Category
Data Exfiltration
Content
| `MINIMAX_API_KEY` | Yes | — | MiniMax API key |
| `WHISPER_MODE` | No | `api` | `api` or `local` |
| `WHISPER_API_KEY` | If mode=api | — | OpenAI API key for Whisper |
| `LLM_BASE_URL` | No | `https://api.minimaxi.com/v1` | MiniMax endpoint |
| `LLM_MODEL` | No | `MiniMax-M2.7` | Model name |

## Supported Sources
Confidence
95% confidence
Finding
The skill is explicitly configured to use an external MiniMax endpoint, and the workflow indicates transcript analysis through that service; in API transcription mode, audio/transcript data may also go to Whisper providers. External transmission is contextually sensitive here because video transcripts can contain personal, proprietary, or regulated information, and the skill does not prominently disclose or constrain that data flow.

Static analysis

No suspicious patterns detected.