# Douyin Video To Doc

> Use when the user provides a Douyin link or share text and wants one-document-per-video output with transcript, source metadata, and a detailed content writeup rather than a short summary

- **Type:** Skill
- **Install:** `agentstack add skill-dolantan548-cmd-douyin-video-to-doc-skill-douyin-video-to-doc-skill`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [dolantan548-cmd](https://agentstack.voostack.com/s/dolantan548-cmd)
- **Installs:** 0
- **Category:** [Content & Media](https://agentstack.voostack.com/c/content-and-media)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [dolantan548-cmd](https://github.com/dolantan548-cmd)
- **Source:** https://github.com/dolantan548-cmd/douyin-video-to-doc-skill

## Install

```sh
agentstack add skill-dolantan548-cmd-douyin-video-to-doc-skill-douyin-video-to-doc-skill
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Douyin Video To Doc

Turn a Douyin share link into a reusable document bundle:

- source video metadata
- local video/audio files
- local transcript files
- one Markdown document per video

Use this skill when the user wants to archive, study, summarize, or structurally organize the full content of a Douyin video.

## Output Rule

Create one Markdown file per video under the configured output directory.

Default output directory:

`%USERPROFILE%\douyin-video-docs`

Override with:

`DOUYIN_VIDEO_TO_DOC_OUTPUT_DIR`

Keep the global skill itself under:

`C:\Users\dolan\.codex\skills\douyin-video-to-doc`

On this machine, if you want outputs under your project workspace root, set:

```powershell
$env:DOUYIN_VIDEO_TO_DOC_OUTPUT_DIR = "D:\dolan_env\temp\project\personal\douyin-video-docs"
```

## Workflow

1. Run the helper script to fetch metadata, download media, extract audio, and generate transcript files.
2. Inspect the transcript and correct obvious ASR mistakes through context.
3. Write a final Markdown document with:
   - basic video info
   - processing notes
   - detailed content writeup
   - transcript raw text
4. Be explicit when transcript text came from local ASR rather than official captions.

## Script

Use:

```powershell
python C:\Users\dolan\.codex\skills\douyin-video-to-doc\scripts\douyin_video_to_doc.py --input "抖音分享文本或链接"
```

Optional flags:

```powershell
python C:\Users\dolan\.codex\skills\douyin-video-to-doc\scripts\douyin_video_to_doc.py --input "..." --model small --output-dir "D:\dolan_env\temp\project\personal\douyin-video-docs"
```

The script produces:

- `VIDEO_ID.mp4`
- `VIDEO_ID.wav`
- `VIDEO_ID..srt`
- `VIDEO_ID..txt`
- `VIDEO_ID.metadata.json`
- `YYYY-MM-DD-douyin--.draft.md`

Example:

```powershell
python C:\Users\dolan\.codex\skills\douyin-video-to-doc\scripts\douyin_video_to_doc.py --input "7.99 复制打开抖音，看看【每日AI评论的作品】简单讲讲最近最火的Hermes Agent https://v.douyin.com/j6ySf9z4GKg/"
```

If you want to pin output to a custom directory for the current shell:

```powershell
$env:DOUYIN_VIDEO_TO_DOC_OUTPUT_DIR = "D:\dolan_env\temp\project\personal\douyin-video-docs"
python C:\Users\dolan\.codex\skills\douyin-video-to-doc\scripts\douyin_video_to_doc.py --input "https://v.douyin.com/xxxx/"
```

## Writing Standard

The final document should not stop at a framework-level summary.

Expand the video into:

- the speaker's main claim
- argument structure
- implementation details
- examples and mechanisms
- assumptions and caveats
- inferred corrections for obvious ASR errors

If a key proper noun is uncertain, say it is inferred from context.

## Verification

Before claiming completion:

- confirm the draft and transcript files exist
- read the transcript text
- make sure the final Markdown includes both transcript and detailed writeup

## Notes

- Douyin pages often do not expose reusable official subtitles.
- In that case the transcript is local ASR output, not platform-native captions.
- Prefer the `small` model when quality matters more than speed.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [dolantan548-cmd](https://github.com/dolantan548-cmd)
- **Source:** [dolantan548-cmd/douyin-video-to-doc-skill](https://github.com/dolantan548-cmd/douyin-video-to-doc-skill)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-dolantan548-cmd-douyin-video-to-doc-skill-douyin-video-to-doc-skill
- Seller: https://agentstack.voostack.com/s/dolantan548-cmd
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.