zzybrim

You can parse Douyin share links to download videos without watermarks, transcribe spoken audio into text, extract body copy from image-text posts, batch-download post images, and archive everything as structured notes inside an Obsidian vault.

It targets content researchers, social media analysts, and knowledge managers who collect Douyin videos or image-text posts and want them organized into an Obsidian-based personal knowledge base with transcribed copy and downloaded media.

The skill requires Python 3.8 or higher and active internet access to reach Douyin. Video speech transcription additionally depends on the SiliconFlow API, so you need valid credentials for that service before extracting spoken-word text.

For video posts it downloads the watermark-free file and transcribes the voiceover into text. For image-text posts it extracts the written body copy and batch-downloads all attached images, then files both formats into your Obsidian vault structure.

The skill itself carries no stated license fee in its manifest, but video transcription consumes SiliconFlow API usage, which may incur costs under that provider's pricing. Douyin access and Obsidian usage follow their own respective terms.