Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add liusencomic-cyber/douyin-agent-kit --skill video-transcribegit clone --depth 1 https://github.com/liusencomic-cyber/douyin-agent-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/liusencomic-cyber/douyin-agent-kit/video-transcribe)<a href="https://agentmods.dev/skills/liusencomic-cyber/douyin-agent-kit/video-transcribe"><img src="https://agentmods.dev/badge/skills/liusencomic-cyber/douyin-agent-kit/video-transcribe/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/liusencomic-cyber/douyin-agent-kit/video-transcribe"><img src="https://agentmods.dev/badge/skills/liusencomic-cyber/douyin-agent-kit/video-transcribe.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00068 | $0.08336 |
| Opus 5 | $0.00034 | $0.04168 |
| Sonnet 5 | $0.00014 | $0.01667 |
| Haiku 4.5 | $0.00007 | $0.00834 |
Grade C, and why
douyin-video-transcribe scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf /tmp/douyin_note_imgs/ Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s -L -o /dev/null -w "%{url_effective}" "https://v.douyin.com/XXXX/" How it starts
The opening of the file, as written. The whole thing — 389 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent 无关声明：本技能与具体 Agent 解耦，可在 Hermes / OpenClaw / WorkBuddy / Claude Code / Codex / 任意本地 Agent 上运行。文中提到的
vision_analyze等工具名 仅为示例：如你的 Agent 无此能力，请按agents/adapter-contract.md适配，视觉/OCR 走douyin_agent_kit.config.yaml的vision.ocr_provider/vision.fallback_ocr。 各 Agent 接入方式见agents/<你的agent>.md（设置向导生成的AGENT_SETUP.md会指向对应文档）。
抖音内容提取与智能归档
将抖音视频/图文 → 提取文字 → 归纳总结 → 判断意图 → 自动执行后续动作。
脚本速查：下载统一用
scripts/douyin_download.py（Scrapling 版，视频+图文双类型，直连→DynamicFetcher→StealthyFetcher 自动降级）。必须用安装了 scrapling 的 Python 运行（系统 Python 可能无 scrapling，需 pip install scrapling）。
环境依赖
| 环节 | 工具 | 说明 |
|---|---|---|
| 浏览器自动化 | Scrapling DynamicFetcher/StealthyFetcher | 访问抖音页面，获取cookie、视频URL、图片 |
| 下载视频/图片 | 页面内fetch() | 利用浏览器cookie绕过防盗链 |
| 提取音频 | ffmpeg |
提取音频轨为 wav |
| 语音转文字 | whisper (homebrew) |
turbo 模型，首次下载 ~1.5GB |
| 视频截帧 | ffmpeg |
每3秒截一帧用于画面OCR |
| 视频画面OCR | 视觉模型 (vision_analyze) | 识别店铺招牌、店名、路标、字幕标注 |
| 图片OCR | 视觉模型 (vision_analyze) | 识别图文作品中的文字 |
| 归纳总结 | LLM | 基于原文生成结构化总结 |
| 意图判断 | LLM | 分析内容类型，决定后续动作 |
| 归档 | Obsidian / Todo | 根据意图写入对应位置 |
工作流程总览
用户发链接 → 解析短链 → 判断类型(视频/图文)
├─ 视频: douyin_download.py下载 → whisper转录 + 截帧OCR
└─ 图文: douyin_download.py下载 → 页面文字 + 图片OCR
→ 合并文字 → LLM归纳总结 → **在聊天中展示完整结果**
├─ 工具类/策略类 → 意图判断 → 归档/待办
└─ 知识/方法论类 → **深度分析（理论映射）** → 展示分析 → 归档
Step 0: Playwright 启动（图文和视频共用）
douyin_download.py 基于 Scrapling 已内置浏览器启动逻辑。若需手动写 Playwright 脚本：
- Playwright 1.59+ 默认用
chromium_headless_shell，安装可能不完整。必须先用find_chromium()探测实际路径（~/Library/Caches/ms-playwright/下Google Chrome for Testing.app或Chromium.app），通过executable_path参数启动。 - 不要浪费时间
playwright install（经常超时）。若 executable_path 不存在，可创建符号链接指向已装 Chrome，但优先 executable_path。 - 详见
references/playwright-chromium-path.md。
Step 1: 解析链接 & 判断类型
短链格式 v.douyin.com/XXXX 先解析重定向获取真实URL：
curl -s -L -o /dev/null -w "%{url_effective}" "https://v.douyin.com/XXXX/"
根据URL路径判断类型：
/video/<ID>→ 视频类型/note/<ID>→ 图文作品- 其他路径 → 按视频流程处理
What ships with it
8 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/blob-video-performance-capture.md 2.0 KB
- references/content-analysis-deep-dive.md 2.9 KB
- references/douyin-page-structure.md 2.2 KB
- references/douyin-shortlink-redirect-pitfall.md 2.7 KB
- references/local-ocr-evaluation.md 1.8 KB
- references/playwright-chromium-path.md 2.0 KB
- references/scrapling-alternative.md 4.4 KB
- scripts/douyin_download.py 13 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 389 lines · 68 tokens per session scan C d4bff768890e
douyin-video-transcribe is a skill published in the GitHub repository liusencomic-cyber/douyin-agent-kit (2 stars, last pushed 10d ago), licensed MIT. It adds 68 tokens to every session and 8,336 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
douyin-video
A tool for downloading videos from Douyin, a Chinese short-video platform, and extracting spoken words as text. It can save downloads and transcripts in folders named after each video.
p5js
Use when users request: p5.js sketches, creative coding, generative art, interactive visualizations, canvas animations, browser-based visual art, data viz, shader effects, or any p5.js project.
audiocraft-audio-generation
AudioCraft: MusicGen text-to-music, AudioGen text-to-sound.
hyperframes
Render MP4/WebM videos from HTML compositions.
ascii-video
ASCII video: convert video/audio to colored ASCII MP4/GIF.
stable-diffusion
Text-to-image generation, inpainting, and img2img.