Boxed FFmpeg allows AI agents to perform secure audio and video processing using a sandboxed FFmpeg WASM component.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install boxed-ffmpeg
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install boxed-ffmpeg using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Boxed FFmpeg?
Boxed FFmpeg is a specialized utility designed for the Openclaw Skills ecosystem that brings the power of FFmpeg to AI coding agents within a secure, isolated WebAssembly (WASM) environment. By utilizing a sandbox, it enables agents to inspect media metadata, convert video formats, and extract audio streams without requiring a native FFmpeg installation on the host system.
This skill ensures that media manipulation tasks are performed locally, preserving privacy and security while providing consistent results across different operating systems. As part of the broader collection of Openclaw Skills, it provides a standardized interface for agents to handle complex multimedia tasks via simple tool calls.
Boxed FFmpeg Use Cases
- Extracting technical metadata and stream information from video or audio files.
- Converting video files between different containers and formats such as MP4 to AVI.
- Stripping audio tracks from video files to create standalone MP3 or AAC files.
- Automating media preparation tasks within a developer's workspace using AI agents.
How Boxed FFmpeg Works
- The AI agent detects a media-related request such as a conversion or information query.
- The system verifies that the required WASM sandbox environment is configured and the component binary is available.
- The agent triggers the wasm-sandbox-run tool, pointing to the specific directory containing the media files.
- The FFmpeg command is executed within the isolated WASM container, ensuring no side effects on the host OS.
- The resulting output file or metadata is returned to the workspace for the user or next agent step.
Boxed FFmpeg Setup
To use this skill, you must first install the required sandbox plugin for Openclaw Skills:
openclaw plugins install clawhub:openclaw-wasm-sandbox
openclaw plugins update openclaw-wasm-sandbox
openclaw gateway restart
Next, ensure the WASM component is downloaded to the correct local directory:
wasm-sandbox-download({
url: "https://raw.githubusercontent.com/guyoung/wasm-sandbox-openclaw-skills/main/boxed-ffmpeg/files/boxed-ffmpeg-component.wasm",
output: "~/.openclaw/skills/boxed-ffmpeg/files/boxed-ffmpeg-component.wasm",
resume: false,
timeout: 120000
})
Boxed FFmpeg Data Schema & Taxonomy
The skill interacts with the local file system using the following parameter structure:
| Parameter | Type | Description |
|---|---|---|
| wasmFile | String | Path to the boxed-ffmpeg-component.wasm file. |
| workDir | String | Absolute path to the directory containing input media. |
| args | Array | The FFmpeg command (info, convert, extract-audio) followed by filenames. |
Note: Within the args array, only filenames should be used, as the context is scoped to the workDir.
Boxed FFmpeg Advanced Features
- Sandboxed Execution: Runs media codecs in a restricted environment to prevent security vulnerabilities associated with complex media parsers.
- Local-First Processing: No cloud dependencies or network access required for processing, making it ideal for private Openclaw Skills deployments.
- Zero-Install FFmpeg: Provides high-level media utilities without requiring the user to manually install and path the FFmpeg binary on their machine.
- Flexible Directory Scoping: Supports deep-directory processing by dynamically setting the workDir for different project structures.
SKILL.md
Loading
Related Openclaw Skills
Excel Power Tool
A comprehensive Excel management and AI-powered analysis tool designed for OpenClaw agents.
volc-docs: Volcengine Documentation Agent
A specialized tool for AI agents to search and retrieve comprehensive official documentation, APIs, and SDK guides from Volcengine.
Local Voice Agent
A complete offline voice-to-voice AI assistant for OpenClaw using Whisper.cpp and Pocket-TTS for 100% local processing.
PySpark to Excel
Effortlessly converts ASCII-style PySpark table outputs into clean Tab-separated text for direct pasting into spreadsheets.
Feishu Bot Creator
A standardized automation workflow for creating, configuring, and publishing Feishu enterprise self-built bots.
Xiaomi MiMo TTS
An advanced Text-to-Speech skill that converts text into natural human voices with granular control over emotions, dialects, and singing styles.
Featured*
