← Back to Skills Marketplace
363
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install noizai-video-translation
Description
Translate and dub videos from one language to another, replacing the original audio with TTS while keeping the video intact.
Usage Guidance
Before installing: 1) Be aware this skill uploads original audio to the Noiz backend when you pass --ref-audio-track (voice cloning) — do not upload sensitive or private recordings unless you trust Noiz and have permission. 2) The SKILL.md requires NOIZ_API_KEY and ffmpeg, but the registry metadata omitted these; ensure you obtain a NOIZ_API_KEY and have ffmpeg installed. 3) The agent may offer to persist the API key to a .env file or call tts.sh config — prefer not to store secrets in plaintext in shared repos; consider using a secure secret store. 4) Verify the referenced youtube-downloader and tts skills come from trusted sources before cloning them into your skills/ folder. 5) If you need higher assurance, inspect the tts skill's code (network calls, where it sends audio) and test the workflow with non-sensitive sample videos first.
Capability Analysis
Type: OpenClaw Skill
Name: noizai-video-translation
Version: 1.0.0
The skill bundle is designed for video translation and dubbing but contains a shell injection vulnerability in the SKILL.md workflow. It instructs the AI agent to execute bash commands using unsanitized user inputs (e.g., VIDEO_URL), which could be exploited to run arbitrary code. Additionally, the instructions suggest writing sensitive credentials (NOIZ_API_KEY) directly to the .env file, and the scripts (replace_audio.sh and srt_to_duck.py) lack explicit input validation for the file paths and parameters they process.
Capability Assessment
Purpose & Capability
The skill's code and SKILL.md match the stated purpose (download subtitles, translate them, generate TTS audio, and replace the video's audio). However, the registry metadata lists no required env vars or binaries while SKILL.md explicitly requires NOIZ_API_KEY and ffmpeg — this mismatch is a documentation/metadata inconsistency that should be corrected.
Instruction Scope
Runtime instructions direct the agent to send the original video's audio (via --ref-audio-track) to the Noiz TTS backend so it can clone the voice and align emotion/duration. That means user audio will be uploaded to a third-party service; the SKILL.md also tells the agent to optionally persist the NOIZ_API_KEY into a project .env or run the tts.sh config helper. These behaviors are coherent with the skill's purpose but raise privacy and secret-handling concerns and should be made explicit to end users.
Install Mechanism
This is an instruction-only skill with two small local scripts. No network downloads or archive extraction are performed by the skill itself, and the scripts perform only local file operations and ffmpeg invocations. That is low-risk from an install mechanism perspective.
Credentials
The SKILL.md requires a NOIZ_API_KEY (for the Noiz TTS backend) and ffmpeg. These are proportionate to a TTS/voice-cloning workflow, but the registry metadata omitted these requirements. Also note the SKILL.md instructs the agent to offer to persist the API key to .env or via tts.sh — storing secrets in plaintext project files has risks and should be optional and documented.
Persistence & Privilege
The skill does not request always:true, does not modify other skills, and does not require system-wide privileges. It does suggest optionally writing the NOIZ_API_KEY to a project .env, which affects repository files but is within expected installer behavior for an API-backed skill.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install noizai-video-translation - After installation, invoke the skill by name or use
/noizai-video-translation - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Version 1.0.0 – Initial release
- Translate and dub videos from one language to another by replacing original audio with TTS while preserving video.
- Supports natural-sounding, timeline-accurate dubbing synchronized with subtitles.
- Utilizes youtube-downloader for video and subtitle extraction and tts skill for dubbed audio generation.
- Allows voice cloning to match original speaker’s tone and emotion for each segment.
- Presents the final dubbed video file for download after processing.
Metadata
Frequently Asked Questions
What is video-translation?
Translate and dub videos from one language to another, replacing the original audio with TTS while keeping the video intact. It is an AI Agent Skill for Claude Code / OpenClaw, with 363 downloads so far.
How do I install video-translation?
Run "/install noizai-video-translation" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is video-translation free?
Yes, video-translation is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does video-translation support?
video-translation is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created video-translation?
It is built and maintained by kusuriuri (@ksuriuri); the current version is v1.0.0.
More Skills