Communitygithub.com

Book2Video

Transforms long books, documents, PDFs, EPUBs, reading materials, classics, social science works, business books, or research papers into viral-ready Chinese video script drafts for platforms like WeChat Video and Douyin, outputting a project-usable raw JSON.

What is Book2Video?

This skill converts lengthy reading materials into engaging short video scripts, targeting Chinese social platforms like WeChat Video (Video Account) and Douyin. It first analyzes the content for suspense, knowledge value, and emotional resonance, rather than simply summarizing chapters. The skill guides the user to define a sharp angle that viewers care about, select a narrative archetype from a set of proven video patterns, and craft a script that feels like a natural conversation, not a lecture. A four-layer quality check ensures the final output is free of forbidden expressions, has good pacing, is well-supported by evidence from the source, and sounds like a real person sharing an interesting insight. The final deliverable is a raw JSON object containing the title, cover copy, key quotes, and the full script, ready for project use. It works with any large text source provided by the user, whether extracted from a PDF, EPUB, or pasted document.

Works with~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/lzcjsyr/Book2Video/tree/HEAD/skills/step1/book-video-script_V2

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

Enhanced Long Book to Video Script Generator

This skill audits long material reading, maintains coverage checklists, performs one-pass revision with auditing, and outputs raw JSON:

  • First, judge whether the content has suspense, knowledge value, and emotional resonance, rather than summarizing by chapter.
  • Define a sharp angle that the audience cares about, then write the script.
  • Use a reading video archetype to organize material, avoiding a book report style.
  • Make knowledge feel like it's naturally shared in a conversation, not a classroom lecture.
  • Use a four-layer quality check to remove forbidden expressions, fix pacing, ensure content support, and add a human touch.

The goal of this skill is to make a book-talking script feel like someone who has thoroughly read the material and understands the average person's situation is earnestly explaining something worth knowing.

File Responsibilities

  • references/reading-strategy.md: Extracting long materials, reading with Bash character windows, maintaining coverage checklists, and a coverage gate.
  • references/writing-standard.md: Angle judgment for book video scripts, narrative archetypes, script requirements, and the final JSON contract.
  • references/revision-workflow.md: Angle brief, first draft, one-pass final revision, and revision audit storage process.
  • references/script-quality-checklist.md: An internal checklist adapted from the four-layer quality check specifically for book-talking scripts.

Body Length Configuration

This is the single source of truth for body draft length in this skill. Other references must not repeat specific word counts but only reference this section.

draft_min_chars: 3000 final_target_chars: 3200

  • _draft_v1.txt must reach draft_min_chars before saving.
  • All subsequent body drafts must not fall below draft_min_chars.
  • The final raw JSON's content target length should follow final_target_chars; total_length should be the actual character count of content.

Main Workflow

Do not output directly in one round. Follow: "Understand Material -> Define Angle -> Select Archetype -> Write First Draft -> Revise and QC Final Draft -> Package JSON"

  1. Material Understanding: First, read references/reading-strategy.md. Extract text, create a Bash character window plan, build and update a coverage checklist.
  2. Coverage Gate: Do not write the angle brief, first draft, revision, or final JSON until the coverage check passes.
  3. Angle Brief: Read references/writing-standard.md, silently complete the HKR (Hook, Knowledge, Resonance) and four major viral gene assessments, select the reading video archetype and propagation anchor, and save _angle_brief.json.
  4. First Draft Writing: Based on _angle_brief.json and writing-standard.md, generate _draft_v1.txt that meets draft_min_chars.
  5. Final Revision: Read references/revision-workflow.md and references/script-quality-checklist.md. Revise _draft_v1.txt into _draft_final.txt in one pass. During revision, perform L1-L4 checks internally. If L1 or L2 fails, do not package the final JSON.
  6. Package Output: From _draft_final.txt, generate the title, cover copy, key quotes, and final raw JSON. Unless the user asks to show the process, only output the project-usable raw JSON, ensuring it can be parsed by json.loads.

Self-Check Requirements

Before submitting the final JSON, silently confirm:

  • That coverage_check.passed=true in the coverage checklist is consistent with the facts, the coverage rate meets the tier requirement, and the beginning, middle, end, and main chapters are covered.
  • That _angle_brief.json explains the viral genes, propagation anchor, core question, audience relationship, counter-intuitive points, reading video archetype, evidence boundaries, and comment controversy points.
  • That the script is not a flat chapter-by-chapter summary or a "what this book says" book report.
  • That there is a cognitive advancement roughly every 250 characters, with frequent returns to the main thread.
  • That important viewpoints are supported by evidence from the source: people, events, scenarios, systems, experiments, data, or original text.
  • That the final revision completed internal checks per the four-layer quality check; L1/L2 must pass, L3/L4 must have no unresolved critical issues; the check summary is written into _revision_audit.json.
  • That the revision process is recorded per revision-workflow.md, the final draft comes from _draft_final.txt, and the audit can account for the real differences from the first draft to the final draft.
  • That the JSON conforms to the final JSON output contract in writing-standard.md.

Individual skills in this repo

This repo contains 6 individual skills — each has its own dedicated page.

lzcjsyr/Book2Video

将 PDF、EPUB、DOCX 等书籍或长文档一次性转成约 4 分钟的中文读书带货短视频口播稿,以情绪共鸣、观看留存和购买欲为核心,直接输出 Book2Video 后续流程可读取的 raw JSON。用户提到读书带货、图书推广、卖书口播、商品卡、走心书单视频、情绪型读书文案或要求生成可直接进入视频流水线的带货稿时使用。

lzcjsyr/Book2Video

将长书、长文档转成中文视频号读书解说稿;先抽取文本并建立可恢复覆盖台账,覆盖自检通过后再按入口定义的正文长度生成初稿、多轮修订和最终 raw JSON。

lzcjsyr/Book2Video

Cody的公众号长文写作skill。当用户需要撰写公众号文章、写稿子、续写文章、根据素材产出长文时使用。触发词包括但不限于:写文章、写稿子、帮我写、续写、扩写、公众号文章、长文、出稿、按我的风格写。即使用户只是说"帮我把这个写成文章"或"用我的风格写一下",只要上下文涉及内容创作和公众号输出,都应该触发。也适用于用户丢过来一个PDF、brief、新闻链接、语音转文字或任何素材说"帮我写篇文章"的场景。不要用于短内容(小红书帖子、推特、朋友圈)或纯标题摘要生成。

lzcjsyr/Book2Video

将中短篇专业报告、公司财报、券商研报、招股书、宏观经济分析及含有较多表格和数据的PDF文档,转化为高品质、数据精准的中文视频号/抖音口播解说稿。只要用户提到“研报”、“财报”、“商业分析报告”或需要“解读数据”、“把报告/PDF做成短视频”,务必使用此 Skill。

lzcjsyr/Book2Video

简化版的文稿生成技能。能够直接读取图书/文档,提取核心论点,生成高品质、口语化的视频号/抖音读书解说文稿并输出标准 JSON 格式。

lzcjsyr/Book2Video

The fallback workflow for authoring custom HyperFrames video compositions at any length or format — longer or multi-scene pieces, brand / sizzle reels, montages, title cards, static loops, and freeform compositions. Input- and length-agnostic. If a specialized workflow clearly fits the input — a marketed product, a website, a topic explainer, a GitHub PR, existing footage, a short motion graphic, or a Remotion port — prefer it (see /hyperframes); use this only as the general fallback when none fit.

Related Skills