video-to-textSAFE
Extract transcripts from Douyin/Kuaishou share links or local video files. 抖音/快手视频转文字。
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Douyin/Kuaishou share link → metadata; local video file → full speech transcript.
抖音/快手分享链接 → 标题/作者/点赞等元数据(无需 Cookie);本地视频 → 语音转写全文。
[](LICENSE) [](#install) [](https://github.com/jiawood2006/hermes-skills)
Install
# 1) Agent Skills CLI (works with 50+ agent hosts) npx skills add jiawood2006/video-to-text # 2) Claude Code / any host that reads ~/.claude/skills git clone https://github.com/jiawood2006/video-to-text.git ~/.claude/skills/video-to-text # 3) Hermes Agent git clone https://github.com/jiawood2006/video-to-text.git ~/.hermes/skills/utilities/video-to-text
No build step, no account, no telemetry.
Usage
# 1) share link → metadata (no login, no cookie) python3 scripts/vtt.py "https://v.douyin.com/xxxx/" # 2) local video → transcript python3 scripts/vtt.py 本地视频.mp4
Ask your agent: "把这个抖音链接的文案提取出来" → it picks the right mode for you.
Requirements
Python 3.9+
pip3 install faster-whisper # transcription only; first run downloads the tiny model (~75MB)
macOS ships avconvert; on Linux install ffmpeg. Share-link parsing needs no dependencies at all.
Notes
Metadata mode is honestly metadata: it does not download or transcribe the video. Transcription runs only on local files. Both paths are self-hosted — nothing is sent to a third-party service.
License
MIT — see LICENSE. Issues and PRs welcome. Part of the hermes-skills collection.
b62e3a14db58OBSERVED · 2026-10-13Install
Commands as the repository documents them. They are shown, not run.
npx skills add jiawood2006/video-to-text
git clone https://github.com/jiawood2006/video-to-text.git ~/.claude/skills/video-to-text
git clone https://github.com/jiawood2006/video-to-text.git ~/.hermes/skills/utilities/video-to-text
Host compatibility
What the documentation claims. We have not run a compatibility test.
| Host | Status | Notes |
|---|---|---|
| claude-code | mentioned |
What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
---
name: video-to-text
description: "视频转文字。用户发抖音分享链接(获取标题/作者/点赞等元数据),或本地视频文件(语音转写全文)时使用。抖音链接用 SSR 解析无需 Cookie。Video to text: extract metadata from Douyin share links, or transcribe speech from local video files."
version: 1.0.0
author: 涛哥
license: MIT
metadata:
hermes:
tags: [video, transcription, douyin, whisper, speech-to-text]
category: utilities
homepage: https://gitee.com/tao6677/useful-tools
---
# Video-to-Text 视频转文字
把视频变成可复制、可搜索的文字。支持抖音分享链接(元数据,无需登录)和本地视频文件(语音转写)。
## 触发条件
用户发送以下内容时使用本技能:
- 抖音分享链接,要求"转文字""提取文案""拆解视频"
- 本地视频文件(.mp4/.mov),要求"语音转文字""字幕"
## 使用步骤
### 1. 抖音分享链接 → 元数据
```bash
python3 ~/.hermes/skills/utilities/video-to-text/scripts/vtt.py "https://v.douyin.com/xxxx/"
```
- 输出元数据(标题/作者/点赞/标签)
- **注意**:分享链接当前只提取元数据——语音转写请让用户保存视频后走本地文件(步骤 2)
### 2. 本地视频文件 → 语音转写
```bash
python3 ~/.hermes/skills/utilities/video-to-text/scripts/vtt.py 本地视频.mp4
```
- 自动提取音频、faster-whisper 转写(中文)
- 转写文本保存为 `<输入名>_transcript.txt` 并打印时间戳分段
## 依赖(首次使用时安装)
```bash
pip3 install faster-whisper # 转写需要(首次会下载 tiny 模型 ~75MB)
# macOS 自带 avconvert;Linux 需 ffmpeg
# 抖音 SSR 解析零依赖(curl + 标准库)
```
## 已知陷阱
- **duration_ms=0 假阴性**:SSR 元数据时长可能为 0,但实际是视频。若元数据显示 0 秒且有下载 URL,先下载验证,不要直接判定为图文。
- **play_addr URL 过期**:SSR 解析出的视频 URL 可能 404,此时需 Cookie 降级或让用户提供文件。
- **whisper 转写慢**:Intel Mac 上 3 分钟视频约 5 分钟;转写必须加 `language='zh'` 否则自动检测多花 1-2 分钟。
- **图文内容**:duration=0 且下载文件 <1MB 时是图文(无音频),只输出元数据,不要编造内容。Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (4 observation(s))
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
b62e3a14db58full audit observations/trust-audit/skill/jiawood2006__video-to-text.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-13 | b62e3a14db58 | SAFE | B | 89 | first audit |
Questions
What does the video-to-text skill do?
Extract transcripts from Douyin/Kuaishou share links or local video files. 抖音/快手视频转文字。
Is video-to-text safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can video-to-text access on my machine?
The audit observed that it reaches the network. Each of those is consistent with what it says it does. Secrets in the source: none found.
Which assistants does video-to-text work with?
Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (b62e3a14db58), read on 2026-10-13. The repository is watched, and a new audit runs when it changes — this is the first audit.