← 返回 Skills 市场
weiqingtangx

Transcribe audio/video to LRC, SRT, WebVTT, ASS, TTML and create karaoke videos

作者 weiqingtangx · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ✓ 安全检测通过
66
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install karadeo
功能描述
Transcribe audio or video files to time-synced lyrics or subtitle formats like LRC, SRT, WebVTT, ASS, and TTML, and create karaoke videos.
使用说明 (SKILL.md)

Karadeo — Agent & AI Integration Guide

Karadeo is a karaoke and lyrics tooling platform. This document describes all available APIs, tools, and agent integration surfaces.

Authentication

All API endpoints require a Bearer API key from https://karadeo.com/dashboard.

Authorization: Bearer kd_\x3Cyour-api-key>

MCP Server

Karadeo exposes a full MCP (Model Context Protocol) server for AI agent tool use.

Endpoint: POST https://karadeo.com/api/mcp
Transport: Streamable HTTP (JSON-RPC 2.0)
Discovery: https://karadeo.com/.well-known/mcp/server-card.json

Initialize

{"jsonrpc":"2.0","id":1,"method":"initialize","params":{}}

List Tools

{"jsonrpc":"2.0","id":2,"method":"tools/list","params":{}}

Call Tool

{
  "jsonrpc": "2.0",
  "id": 3,
  "method": "tools/call",
  "params": {
    "name": "transcribe",
    "arguments": {
      "audio_url": "https://example.com/song.mp3",
      "format": "lrc"
    }
  }
}

REST API

Transcribe

Convert audio or video to time-synced lyrics or subtitle files.

Endpoint: POST https://karadeo.com/api/transcribe
Docs: https://karadeo.com/resources/karadeo-lyrics-api
OpenAPI: https://karadeo.com/api/doc

Request body (JSON):

Field Type Required Description
fileUrl string yes Publicly accessible audio/video URL
format string no lrc, srt, ass, webvtt, ttml, txt (default: lrc)
transcriptText string no Known lyrics to align instead of auto-transcribing
isWordLevel boolean no Word-level timing (default: false)

Example:

curl -X POST https://karadeo.com/api/transcribe \
  -H "Authorization: Bearer kd_your_key" \
  -H "Content-Type: application/json" \
  -d '{"fileUrl":"https://example.com/song.mp3","format":"lrc"}'

Response: Plain text subtitle file with appropriate Content-Type header.


WebMCP (Browser)

Karadeo registers browser tools via navigator.modelContext.registerTool() on every page load.

Available tools:

  • list-karaoke-tools — list all Karadeo tools with URLs
  • navigate-to-tool — navigate to a specific tool page
  • list-karaoke-templates — list available karaoke video templates

Discovery Endpoints

Endpoint Description
/.well-known/agent-skills/index.json Agent Skills Discovery index (RFC v0.2.0)
/.well-known/mcp/server-card.json MCP Server Card (SEP-1649)
/.well-known/api-catalog API Catalog (RFC 9727)
/.well-known/oauth-authorization-server OAuth discovery metadata (RFC 8414)

Content Policy

Content-Signal: ai-train=no, search=yes, ai-input=no

AI agents may read this site for tool use. Training on site content is not permitted.

安全使用建议
This skill appears coherent and purpose-aligned. Before installing or using it, confirm you trust karadeo.com, protect your API key, and avoid submitting private audio/video URLs unless you are comfortable sharing them with the service.
功能分析
Type: OpenClaw Skill Name: karadeo Version: 1.0.1 The skill bundle provides documentation and API specifications for Karadeo, a karaoke and lyrics transcription service. It defines standard Model Context Protocol (MCP) and REST API endpoints hosted on karadeo.com. There are no signs of malicious code, data exfiltration, or prompt injection attempts; the content is strictly informational for agent integration.
能力标签
requires-oauth-tokenrequires-sensitive-credentials
能力评估
Purpose & Capability
The documented REST API and MCP example are aligned with transcribing audio/video into subtitle or lyric formats. No unrelated local access or destructive behavior is shown.
Instruction Scope
The skill documents remote REST, MCP, and browser WebMCP tool surfaces. These are purpose-aligned, but users should understand that an agent may interact with Karadeo-hosted tools.
Install Mechanism
There is no install spec, no code files, no required binaries, and no static scan findings.
Credentials
The workflow depends on sending publicly accessible audio/video URLs, and optionally transcript text, to Karadeo. This is expected for transcription but matters for private media.
Persistence & Privilege
Karadeo API use requires a bearer API key. No local persistence, background process, or privileged system access is described.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install karadeo
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /karadeo 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.1
- No user-visible changes; documentation and skill functionality remain unchanged in this version.
v1.0.0
Karadeo-transcribe 1.0.0 — Initial release - Introduces detailed integration guide for agents and AI, including authentication instructions. - Documents full MCP (Model Context Protocol) server endpoints and usage examples. - Provides REST API reference for audio/video transcription to time-synced lyrics or subtitles in multiple formats. - Outlines browser tool registration and capabilities via WebMCP. - Lists discovery endpoints for skill, API, and OAuth metadata. - States content policy restricting AI training and input usage.
元数据
Slug karadeo
版本 1.0.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 2
常见问题

Transcribe audio/video to LRC, SRT, WebVTT, ASS, TTML and create karaoke videos 是什么?

Transcribe audio or video files to time-synced lyrics or subtitle formats like LRC, SRT, WebVTT, ASS, and TTML, and create karaoke videos. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 66 次。

如何安装 Transcribe audio/video to LRC, SRT, WebVTT, ASS, TTML and create karaoke videos?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install karadeo」即可一键安装,无需额外配置。

Transcribe audio/video to LRC, SRT, WebVTT, ASS, TTML and create karaoke videos 是免费的吗?

是的,Transcribe audio/video to LRC, SRT, WebVTT, ASS, TTML and create karaoke videos 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Transcribe audio/video to LRC, SRT, WebVTT, ASS, TTML and create karaoke videos 支持哪些平台?

Transcribe audio/video to LRC, SRT, WebVTT, ASS, TTML and create karaoke videos 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Transcribe audio/video to LRC, SRT, WebVTT, ASS, TTML and create karaoke videos?

由 weiqingtangx(@weiqingtangx)开发并维护,当前版本 v1.0.1。

💬 留言讨论