youtube-to-docs

Click on the image below to see a demo of YouTube to docs:

Convert YouTube videos into structured docs, summaries, audio, and visual assets for easier discovery.

View all available CLI options:

uvx youtube-to-docs --help

To keep the installation light, some features are optional. You can enable them by specifying "extras":

audio: Required for TTS and audio processing (uses yt-dlp).
video: Required for video generation (uses static-ffmpeg).
workspace: Required for Google Drive integration.
m365: Required for Microsoft SharePoint/OneDrive integration.
aws: AWS Bedrock support.
azure: Required for Azure OpenAI models.
gcp: Required for Google Gemini, Vertex AI models, and GCP Cloud TTS (uses google-genai and google-cloud libs).
app: Web app with browser-based UI (uses fastapi, uvicorn).
all: Installs everything.

Example: Run with audio and video support

uvx --with "youtube-to-docs[audio,video]" youtube-to-docs ...

Example: Run with everything

uvx --with "youtube-to-docs[all]" youtube-to-docs ...

Note: The commands above require uv. You can install it via:

macOS/Linux: curl -LsSf https://astral.sh/uv/install.sh | sh
Windows: powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Install as a Gemini CLI extension:

gemini extensions install https://github.com/DoIT-Artificial-Intelligence/youtube-to-docs.git

Install as a Claude Code plugin:

/plugin marketplace add DoIT-Artificial-Intelligence/youtube-to-docs
/plugin install youtube-to-docs@youtube-to-docs

Run as a web app:

uvx --from "youtube-to-docs[all]" youtube-to-docs-app

Then open http://localhost:8000 in your browser. See the Web App docs for details.

Install as an MCP server for Claude Code:

claude mcp add youtube-to-docs -- uvx --with "youtube-to-docs[all]" python -m youtube_to_docs.mcp_server

Created with the help of AI. All artifacts have been checked and work as expected.

What's Inside