AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
翻译准备中
尚未检测到更新。GitHub 发布版本后会显示在这里。
个人 GitHub 账号下公开仓库的所有者可以登录并验证所有权,自行管理介绍,无需管理员批准。
登录并生成验证文件,在 30 分钟内将其提交到此仓库的默认分支。其他分支或分叉仓库中的文件无效。然后在下方检查文件,即可无需管理员批准直接管理项目。请保留文件以维持管理权限。
如需登录,将前往登录页面。登录后请回到此处,再次点击所选的验证按钮。
在项目文件夹中运行 /nomorevibe verify,即可验证所有权并自行管理介绍。
一分钟完成验证YouTube Transcript API clients (JavaScript, Python) and examples for TranscriptYT — transcripts as JSON, text, SRT, VTT in 150+ languages
翻译准备中
取自仓库 README
正在加载完整 README。 以下为已保存的摘录。
TranscriptYT: YouTube Transcript API clients and examples
Official JavaScript and Python clients for the TranscriptYT YouTube Transcript API, plus examples.
Get the transcript of any public YouTube video as timestamped JSON, plain text, SRT, VTT, CSV, or Markdown, in 150+ languages. Videos without captions are transcribed with AI. Free tier: 100 transcripts, no card. Get an API key →
• API docs · OpenAPI spec · Pricing
• MCP server for Claude Code, Claude Desktop, Cursor, Codex, and VS Code
• Free web tool (no code)
See js/README.md.
See python/README.md.
Use it from an AI agent (MCP)查看完整 README ↗Official JavaScript and Python clients for the TranscriptYT YouTube Transcript API, plus examples.
Get the transcript of any public YouTube video as timestamped JSON, plain text, SRT, VTT, CSV, or Markdown, in 150+ languages. Videos without captions are transcribed with AI. Free tier: 100 transcripts, no card. Get an API key →
npm install transcriptyt
import { TranscriptYT } from "transcriptyt";
const client = new TranscriptYT({ apiKey: process.env.TRANSCRIPTYT_API_KEY });
const transcript = await client.getTranscript("https://youtu.be/dQw4w9WgXcQ");
console.log(transcript.title, transcript.segments.length);
const srt = await client.getTranscript("dQw4w9WgXcQ", { format: "srt" });
See js/README.md.
pip install transcriptyt
from transcriptyt import TranscriptYT
client = TranscriptYT() # reads TRANSCRIPTYT_API_KEY
transcript = client.get_transcript("https://youtu.be/dQw4w9WgXcQ")
print(transcript["title"], len(transcript["segments"]))
srt = client.get_transcript("dQw4w9WgXcQ", format="srt")
See python/README.md.
curl "https://transcript-yt.com/v1/transcript?url=dQw4w9WgXcQ&format=text" \
-H "Authorization: Bearer $TRANSCRIPTYT_API_KEY"
claude mcp add --transport http transcriptyt https://transcript-yt.com/mcp \
--header "Authorization: Bearer $TRANSCRIPTYT_API_KEY"
Then paste a YouTube link into the chat. Setup for Cline, Cursor, and other clients: llms-install.md and the docs. Also on the official MCP Registry as io.github.rajdeep-automation/transcriptyt.
Errors raise TranscriptYTError with status and code. Retry BLOCKED, RATE_LIMITED, and UPSTREAM_ERROR with backoff; VIDEO_UNAVAILABLE, NO_CAPTIONS, and LANGUAGE_NOT_AVAILABLE are final; QUOTA_EXCEEDED (HTTP 402) means you're out of credits. Failed requests are never billed. Full list.
MIT