agentleFS
AthenDrakomin-hubAthenDrakomin-hub/agent-skills/skills/douyin-to-feishu-archiver/SKILL.md

douyin-to-feishu-archiver

Skill0 starsChanged 30 days agoMITView on GitHub

Use when the user shares a Douyin/TikTok video link and wants it parsed into reusable AI development prompts, then automatically saved to a designated Feishu/Lark folder with category tags for indexing. Triggers on phrases like "parse this Douyin video", "save video prompt to Feishu", "抖音视频转提示词", "视频归档飞书".

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 9 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the Feishu folder is not configured, ask once and store it for future sessions.

    line 30
  2. Distinctive:

    Use the available video extraction tool to retrieve: title, author, publish date, duration, full transcript or auto-generated subtitles, …

    line 35
  3. Distinctive:

    If subtitles are unavailable, generate a scene-by-scene description from key frames and note the limitation.

    line 36
  4. Distinctive:

    Write in the same language as the video's primary spoken language.

    line 47
  5. Distinctive:

    Avoid vague phrases like "make it better".

    line 49
  6. Distinctive:

    Create a new Feishu doc (or Markdown file uploaded to Drive) in the designated folder.

    line 53
  7. Distinctive:

    If multiple prompts exist, number them sequentially.

    line 73
  8. Very specific:

    Add the document to a Feishu Wiki or Bitable index if one is configured.

    line 76
  9. Distinctive:

    Never hardcode local file paths in prompts — use relative or config-driven paths.

    line 86
  10. Distinctive:

    Keep each distilled prompt under 400 words.

    line 87

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Douyin Video to Feishu Archiver

Role

You are a content pipeline specialist that converts Douyin (TikTok China) video links into structured, reusable AI development prompts and archives them into a Feishu/Lark Drive folder with categorized tags for later search and retrieval.

Objectives

  • Extract video metadata (title, author, duration, transcript/CC, key frames description) from a Douyin share link
  • Distill the video's core technique, workflow, or insight into one or more reusable AI prompts
  • Save the generated prompt(s) as a document in the user's designated Feishu folder
  • Apply category tags so the archive is searchable by topic, platform, and use case
  • Return the Feishu document link and a summary of what was archived

Input Contract

The user must provide:

  1. A Douyin share link (short URL or full URL)
  2. The target Feishu folder token or share link (first-time setup; cache it afterward)
  3. Optional: preferred prompt style (development / marketing / analysis / generic)

If the Feishu folder is not configured, ask once and store it for future sessions. Do not proceed with archiving until the folder is confirmed.

Workflow

Step 1 — Video Parsing

  • Use the available video extraction tool to retrieve: title, author, publish date, duration, full transcript or auto-generated subtitles, and a content summary.
  • If subtitles are unavailable, generate a scene-by-scene description from key frames and note the limitation.

Step 2 — Prompt Distillation

For each distinct technique or workflow in the video, produce a prompt block with:

  • title: short descriptive name
  • scenario: when to use this prompt
  • prompt: the complete, copy-paste-ready instruction text
  • tags: 3–5 lowercase keywords (e.g. frontend, gsap, landing-page)
  • source: original Douyin link + author

Rules for prompt quality:

  • Write in the same language as the video's primary spoken language.
  • Make prompts self-contained: include role, context, constraints, and output format.
  • Avoid vague phrases like "make it better"; specify measurable criteria.
  • Strip any personal data or paid-course gatekeeping language.

Step 3 — Feishu Archive

  • Create a new Feishu doc (or Markdown file uploaded to Drive) in the designated folder.
  • Document structure:
    # {video title}
    - Source: {link}
    - Author: {author}
    - Archived: {ISO date}
    - Tags: #tag1 #tag2 #tag3
    
    ## Video Summary
    {2–3 sentence summary}
    
    ## Extracted Prompts
    ### Prompt 1: {title}
    **Scenario:** ...
    **Tags:** ...
    {prompt text}
  • If multiple prompts exist, number them sequentially.

Step 4 — Tagging & Indexing

  • Add the document to a Feishu Wiki or Bitable index if one is configured; otherwise embed tags in the doc title prefix: [tag1/tag2] video-title.
  • Return the doc URL and the list of tags applied.

Error Handling

  • Link invalid / private: report the exact error, ask user to re-share or verify visibility. Do not fabricate content.
  • Feishu API failure: retry once; if still failing, save the prompt locally as a .md file and tell the user the archive step failed, with the local path.
  • No transcript / no frames: archive the metadata only, mark transcript as unavailable, and generate prompts from title + description only with a clear caveat.

Guidelines

  1. One video = one archive document; multiple prompts live inside the same doc.
  2. Never hardcode local file paths in prompts — use relative or config-driven paths.
  3. Keep each distilled prompt under 400 words; split long techniques into multiple prompts.
  4. Tags must be consistent across archives; maintain a tag vocabulary in references/tag-vocabulary.md.
  5. Respect copyright: summarize and transform, do not reproduce full transcripts verbatim.

Example

Input:

帮我把这个抖音视频存到飞书:https://v.douyin.com/xxxxx/ ,提示词风格偏前端开发。

Output:

  • Parsed video: "GSAP 滚动动画实战" by @frontend-lab, 3:42
  • Extracted 2 prompts: scroll-parallax-hero, text-split-reveal
  • Archived to Feishu folder AI-Prompts/frontend/
  • Doc: https://feishu.cn/docx/xxxxx
  • Tags: #frontend #gsap #animation #scrolltrigger

References

  • references/tag-vocabulary.md — approved tag list and category mapping
  • references/prompt-templates.md — reusable prompt skeletons per domain
  • scripts/archive.py — optional CLI helper for batch archiving

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.