← Back to Skills Marketplace
volc-ai-mediakit

Byted Mediakit Tools

by Volc-AI-MediaKit · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
214
Downloads
2
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install byted-mediakit-tools
Description
火山引擎 AI MediaKit 音视频处理工具集,提供视频理解、音频提取、视频剪辑、音视频拼接、画质增强、文生视频、音视频合成等能力。当用户提及音频剪辑、视频剪辑、音视频拼接、文生视频、音频提取、画质增强、视频理解、音视频合成、媒体裁剪等需求时必须调用本Skill。当用户需要视频理解时,宿主agent必须自动解...
README (SKILL.md)

说明:宿主若在环境中注入 ARK_SKILL_API_BASE / ARK_SKILL_API_KEY(例如供其他 Skill 走 SkillHub 网关),与本 Skill 的 AMK_API_KEYARK_API_KEY(视频理解)相互独立,请勿混淆。

⚠️ 严格执行:必须先完成 环境检查;环境缺失须提示用户,不可跳过。

\x3CSKILL_DIR>byted-mediakit-tools 所在目录。 当前方法返回的 链接仅供下载,不支持播放能力 禁止修改任何返回数据信息,如 play_urlrequest_idtask_id 等 用户明确声明需要重新执行时:除 understand_video_content 外的方法需 生成新的 client_token(不要复用上一次的 client_token,避免命中上次的幂等结果

火山引擎 AI MediaKit 音视频处理工具集

概览

本工具集基于火山引擎 AI MediaKit 提供一站式音视频处理能力,包括:

  • 🎬 视频理解:AI 分析视频内容,生成自然语言描述
  • ✂️ 音视频剪辑:精确裁剪音视频时长
  • 🔗 音视频拼接:拼接多个片段,支持转场效果
  • 🎵 音频提取:从视频中提取音频轨道
  • 🖼️ 画质增强:提升视频画质、分辨率、帧率
  • 🎬 文生视频:图片生成视频,支持动画效果
  • 🎧 音视频合成:合成新的音轨与视频

获取密钥

在开始使用前,请先获取必要的 API 密钥:

  • AI MediaKit 控制台:https://console.volcengine.com/imp/ai-mediakit/
  • 方舟模型与密钥:https://console.volcengine.com/ark/region:ark+cn-beijing/model/detail?Id=doubao-seed-1-8

快速开始

1. 环境配置

\x3CSKILL_DIR>/.env 中配置环境变量(首次使用会自动创建模板):

# AMK API Key (必填) - https://console.volcengine.com/imp/ai-mediakit/
AMK_API_KEY=your_amk_api_key_here
# AMK 环境取值 prod 或 boe
AMK_ENV=prod
# 是否启用 client_token 自动注入(用于幂等)
AMK_ENABLE_CLIENT_TOKEN=false
# 方舟 密钥(可选,仅使用视频理解功能时必须配置)
ARK_API_KEY=your_ark_api_key_here
# 方舟 模型ID(可选,仅使用视频理解功能时必须配置)
ARK_MODEL_ID=doubao-seed-1-8

2. 依赖安装

cd \x3CSKILL_DIR>/scripts
python3 -m venv .venv
. .venv/bin/activate
python -m pip install -U pip
python -m pip install -r requirements.txt

核心功能

同步能力(立即返回结果)

能力 说明
understand_video_content 视频内容理解,使用 AI 分析视频并生成自然语言描述

异步能力(默认自动等待结果)

能力 说明
trim_media_duration 裁剪音视频时长,精确到毫秒
concat_media_segments 拼接多个音视频片段,支持转场效果
extract_audio 从视频中提取音频轨道,支持 mp3/m4a 格式
enhance_video 视频画质增强,支持超分、插帧等
image_to_video 图片生成视频,支持动画和转场
mux_audio_video 音视频合成,支持时长对齐

辅助能力

能力 说明
query_task 查询异步任务执行状态和结果

使用示例

视频理解

./byted-mediakit-tools.sh understand_video_content \
  --video_url "https://example.com/video.mp4" \
  --prompt "总结视频内容" \
  --fps 1

视频裁剪

# 裁剪前 10 秒
./byted-mediakit-tools.sh trim_media_duration \
  --type video \
  --source "https://example.com/video.mp4" \
  --start_time 0 \
  --end_time 10

音视频拼接

./byted-mediakit-tools.sh concat_media_segments \
  --type video \
  --sources "https://example.com/1.mp4" "https://example.com/2.mp4"

音频提取

./byted-mediakit-tools.sh extract_audio \
  --video_url "https://example.com/video.mp4" \
  --format mp3

画质增强

./byted-mediakit-tools.sh enhance_video \
  --video_url "https://example.com/video.mp4" \
  --tool_version professional \
  --resolution 1080p

图片生成视频

./byted-mediakit-tools.sh image_to_video \
  --images "image_url=https://example.com/1.jpg,duration=3,animation_type=zoom_in" \
           "image_url=https://example.com/2.jpg,duration=3,animation_type=pan_left"

音视频合成

./byted-mediakit-tools.sh mux_audio_video \
  --video_url "https://example.com/video.mp4" \
  --audio_url "https://example.com/audio.mp3" \
  --is_audio_reserve false

异步任务(不等待结果)

# 使用 --no-wait 立即返回 task_id
./byted-mediakit-tools.sh --no-wait trim_media_duration \
  --type video \
  --source "https://example.com/video.mp4" \
  --start_time 0 \
  --end_time 10

# 查询任务结果
./byted-mediakit-tools.sh query_task --task_id "amk-xxx-xxx"

响应格式

同步响应(视频理解)

{
  "status": "success",
  "result": {
    "choices": [
      {
        "role": "assistant",
        "content": "视频内容分析结果..."
      }
    ]
  }
}

异步响应(默认自动等待)

{
  "task_id": "amk-tool-extract-audio-xxxxxxxxxxxxxx",
  "duration": 82.454056,
  "play_url": "https://example.vod.cn-north-1.volcvideo.com/xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx.mp3?preview=1&auth_key=***",
  "request_id": "20260401xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
  "status": "completed",
  "task_type": "extract-audio"
}

异步响应(--no-wait)

{
  "status": "pending",
  "task_id": "amk-xxx-xxx",
  "message": "任务已提交,已跳过等待,可调用 query_task 接口传入 task_id 查询结果",
  "query_example": "./byted-mediakit-tools.sh query_task --task_id amk-xxx-xxx"
}

错误响应

{
  "status": "failed/canceled/timeout",
  "task_id": "amk-xxx-xxx",
  "message": "错误详情"
}

详细文档

各功能的详细参数说明请参考 reference/ 目录下的对应文档:

能力 文档链接
understand_video_content reference/understand_video_content.md
query_task reference/query_task.md
concat_media_segments reference/concat_media_segments.md
enhance_video reference/enhance_video.md
extract_audio reference/extract_audio.md
image_to_video reference/image_to_video.md
mux_audio_video reference/mux_audio_video.md
trim_media_duration reference/trim_media_duration.md
统一响应格式 reference/common_response.md

注意事项

  1. 返回链接:所有返回的 play_url 等链接仅供下载,不支持直接播放
  2. 幂等性:重新执行任务时,请确保生成新的 client_tokenAMK_ENABLE_CLIENT_TOKEN=true 时自动处理)
  3. 视频理解:使用视频理解功能必须配置 ARK_API_KEYARK_MODEL_ID
  4. 超时处理:大文件处理可能耗时较长,建议使用 --no-wait 配合 query_task 轮询

© 北京火山引擎科技有限公司 2026 版权所有

Usage Guidance
What to consider before installing/use: - This skill will require a MediaKit API key (AMK_API_KEY) and AMK_ENV; video-understanding features additionally require an Ark key and model id. Only provide keys with the minimum privileges and ideally create a dedicated key for this purpose. - The skill will ask (or instruct the agent to ask) you to paste keys into a .env file at the repository root; that file is written to disk. Keep secrets out of public repos and avoid pasting credentials into chats. - The bundle contains Python code and a requirements.txt; you must create a virtualenv and pip-install dependencies locally to run it. Inspect the code and run in an isolated environment if you are concerned about side effects. - Network and file read/write permissions are required for normal operation (upload/submit URLs, poll results, save temporary data). If you run this inside an automated agent, be aware that the agent may call the skill autonomously and use any API key available in its environment. - If you only need specific functions (e.g., audio extraction), consider limiting which features you enable and use a scoped API key where possible. Overall: the package appears internally coherent with its stated purpose; no evidence in the provided files of unrelated exfiltration or hidden endpoints. If you want higher assurance, review the included Python files (amk_client/* and main.py) and run the tool in an isolated/test account with restricted API keys first.
Capability Analysis
Type: OpenClaw Skill Name: byted-mediakit-tools Version: 1.0.0 The skill bundle is a legitimate integration for Volcano Engine's AI MediaKit and Ark APIs, providing video understanding, editing, and synthesis capabilities. The code is well-engineered, utilizing Pydantic for robust input validation and the httpx library for asynchronous communication with official ByteDance/Volcano Engine endpoints (e.g., volces.com). No evidence of data exfiltration, unauthorized execution, or malicious prompt injection was found; the requested permissions are consistent with the tool's media processing functions.
Capability Assessment
Purpose & Capability
Name/description (火山引擎 AI MediaKit audio/video tools) matches the included code and docs. The skill implements media operations (trim, concat, extract audio, enhance, image→video, video understanding). Required env vars (AMK_API_KEY, AMK_ENV, optional ARK_API_KEY/ARK_MODEL_ID) are appropriate for the stated integrations.
Instruction Scope
SKILL.md and the code require network, file_read/file_write and temp_storage (to create .env, save temporary state, and to read/write task results). The skill explicitly instructs the host/agent to obtain API keys from the user and write them into a repo-root .env (and will auto-create a template .env). This is expected for a local CLI-based client but means the agent or user will be asked to provide secrets and the skill will write them to disk — treat that as sensitive behavior.
Install Mechanism
No platform install spec is present (instruction-only from registry), but the bundle includes Python code and a requirements.txt. SKILL.md instructs creating a virtualenv and pip installing dependencies. There's no remote download or obfuscated installer in the bundle, but using the skill requires installing third‑party Python packages locally.
Credentials
Requested environment variables (AMK_API_KEY mandatory, AMK_ENV mandatory, AMK_ENABLE_CLIENT_TOKEN optional, ARK_API_KEY/ARK_MODEL_ID optional) are proportional to the functionality. The code only uses these keys to construct Authorization headers for AMK and optional Ark combined mode; no unrelated credentials are requested.
Persistence & Privilege
The skill does not request always:true or any elevated platform persistence. It will read/write a local .env and may create a virtualenv per installation instructions, which is normal for a CLI client. It does not modify other skills or system-wide agent settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install byted-mediakit-tools
  3. After installation, invoke the skill by name or use /byted-mediakit-tools
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
byted-mediakit-tools v1.0.0 - 首发音视频处理工具集,支持火山引擎 AI MediaKit 全系列能力 - 支持视频理解、音频提取、视频/音频剪辑、拼接、画质增强、文生视频、音视频合成等高阶媒体处理 - 完备的环境变量配置方式,详细 API 密钥与参数指南 - 支持同步与异步调用,异步任务支持主动查询 - 明确的响应格式规范与详细错误提示 - 示例命令与参数详尽,便于上手
Metadata
Slug byted-mediakit-tools
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Byted Mediakit Tools?

火山引擎 AI MediaKit 音视频处理工具集,提供视频理解、音频提取、视频剪辑、音视频拼接、画质增强、文生视频、音视频合成等能力。当用户提及音频剪辑、视频剪辑、音视频拼接、文生视频、音频提取、画质增强、视频理解、音视频合成、媒体裁剪等需求时必须调用本Skill。当用户需要视频理解时,宿主agent必须自动解... It is an AI Agent Skill for Claude Code / OpenClaw, with 214 downloads so far.

How do I install Byted Mediakit Tools?

Run "/install byted-mediakit-tools" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Byted Mediakit Tools free?

Yes, Byted Mediakit Tools is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Byted Mediakit Tools support?

Byted Mediakit Tools is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Byted Mediakit Tools?

It is built and maintained by Volc-AI-MediaKit (@volc-ai-mediakit); the current version is v1.0.0.

💬 Comments