Idea To Blueprint
sickn33/agentic-awesome-skills
Turn a raw product, app, bot or feature idea into one evidence-backed build blueprint (researched stack, epics, Given/When/Then criteria, tests) that coding agents build one epic per session.
From an idea to a sellable series design; series engine / franchise, story pattern, character webs that keep generating conflict, pilot types, and the documents that sell them…
$ npx skills add jtydhr88/screenwriting-skills --skill sw-series-engine-bible -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install jtydhr88/screenwriting-skills sw-series-engine-bible --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/jtydhr88/screenwriting-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/screenwriting/skills/sw-series-engine-bible .claude/skills/sw-series-engine-bible && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "sw-series-engine-bible" agent skill from https://github.com/jtydhr88/screenwriting-skills/tree/main/plugins/screenwriting/skills/sw-series-engine-bible into .claude/skills/sw-series-engine-bible/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sw-series-engine-bible", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/jtydhr88/screenwriting-skills/tree/main/plugins/screenwriting/skills/sw-series-engine-bibleType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add jtydhr88/screenwriting-skills --skill sw-series-engine-bible -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install jtydhr88/screenwriting-skills sw-series-engine-bible --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/jtydhr88/screenwriting-skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/plugins/screenwriting/skills/sw-series-engine-bible .agents/skills/sw-series-engine-bible && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "sw-series-engine-bible" agent skill from https://github.com/jtydhr88/screenwriting-skills/tree/main/plugins/screenwriting/skills/sw-series-engine-bible into .agents/skills/sw-series-engine-bible/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sw-series-engine-bible", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add jtydhr88/screenwriting-skills --skill sw-series-engine-bible -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install jtydhr88/screenwriting-skills sw-series-engine-bible --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/jtydhr88/screenwriting-skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/plugins/screenwriting/skills/sw-series-engine-bible .cursor/skills/sw-series-engine-bible && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "sw-series-engine-bible" agent skill from https://github.com/jtydhr88/screenwriting-skills/tree/main/plugins/screenwriting/skills/sw-series-engine-bible into .cursor/skills/sw-series-engine-bible/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sw-series-engine-bible", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/jtydhr88/screenwriting-skills.git --path plugins/screenwriting/skills/sw-series-engine-bible--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add jtydhr88/screenwriting-skills --skill sw-series-engine-bible -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install jtydhr88/screenwriting-skills sw-series-engine-bible --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/jtydhr88/screenwriting-skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/plugins/screenwriting/skills/sw-series-engine-bible .gemini/skills/sw-series-engine-bible && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "sw-series-engine-bible" agent skill from https://github.com/jtydhr88/screenwriting-skills/tree/main/plugins/screenwriting/skills/sw-series-engine-bible into .gemini/skills/sw-series-engine-bible/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sw-series-engine-bible", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install jtydhr88/screenwriting-skills sw-series-engine-bibleInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add jtydhr88/screenwriting-skills --skill sw-series-engine-bible -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/jtydhr88/screenwriting-skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/plugins/screenwriting/skills/sw-series-engine-bible .github/skills/sw-series-engine-bible && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "sw-series-engine-bible" agent skill from https://github.com/jtydhr88/screenwriting-skills/tree/main/plugins/screenwriting/skills/sw-series-engine-bible into .github/skills/sw-series-engine-bible/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sw-series-engine-bible", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add jtydhr88/screenwriting-skills --skill sw-series-engine-bible -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install jtydhr88/screenwriting-skills sw-series-engine-bible --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/jtydhr88/screenwriting-skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/plugins/screenwriting/skills/sw-series-engine-bible .opencode/skills/sw-series-engine-bible && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "sw-series-engine-bible" agent skill from https://github.com/jtydhr88/screenwriting-skills/tree/main/plugins/screenwriting/skills/sw-series-engine-bible into .opencode/skills/sw-series-engine-bible/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sw-series-engine-bible", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
sw-series-engine-bibleFrom an idea to a sellable series design; series engine / franchise, story pattern, character webs that keep generating conflict, pilot types, and the documents that sell them…
Sw Series Engine Bible is an agent skill from jtydhr88/screenwriting-skills. From an idea to a sellable series design; series engine / franchise, story pattern, character webs that keep generating conflict, pilot types, and the documents that sell them (剧集引擎/franchise/故事模式/人物网/试播集类型/剧集圣经/跳板/提案). Merged from Rabkin's four-element franchise and story landmines, Landau's tacit contract and pitch recipe, Oberg's series types and bible specs, Douglas's franchise-plus-springboard and bible tiers, Goldberg & Rabkin's 100-episode test, Blum's concept/presentation/bible ladder, Miller's…
Its SKILL.md is about 6.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files (for example `reference-documents.md`, `reference-samples.md` and `reference.md`).
The repository describes itself as: Professional agent skills for screenwriting, television writing and dramaturgy. The licence is MIT.
10 steps, taken from the first numbered list in SKILL.md.
Read from SKILL.md and the folder at commit 51115f1. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Sw Series Engine Bible loads about 6.7k tokens when it runs. Until then it costs about 249 tokens; SKILL.md has 2,292 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from jtydhr88/screenwriting-skills at commit 51115f1, republished under its MIT licence (© jtydhr88). 2,292 words, ~6,654 tokens.
.claude/skills/sw-series-engine-bible/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.输出语言=提问语言;术语一律锚回原词,见 sw-workflow/terms.md,不自创译名。
管四层:引擎(凭什么每周都有故事)、人物网(冲突从哪里自动产生)、pilot 类型与策略、文档(logline → 提案 → treatment → bible)。
不管:单集与季的幕结构、页码、act out 与 cliffhanger 类型学(→ sw-series-structure);编剧室流程与 pitch 会议礼仪(→ sw-writers-room);半小时喜剧的笑点力学与人物配置(→ sw-sitcom-comedy);国产剧策划书、分集大纲、集数体量与审查(→ sw-chinese-series-practice)。前提/主控思想/三维人物/对白等媒介无关内容不在此重写(→ sw-premise-theme、sw-character-conflict)。
参考资料分三个文件:引擎拆解、剧集类型与人物网见 reference.md;文档模板的完整字段表(Rabkin 十节 pitch document、Smith series format 十一节、Oberg bible 十二组件)见 reference-documents.md;填好的样例(两份 writers’ guidelines、Miller 六份 treatment 矩阵、Succession 七个 want 表)见 reference-samples.md。
本节 44 KB,规则编到 56 条,逐条读完通常是浪费。
| 任务 | 读 | 跳过 |
|---|---|---|
| 判断一个点子能不能做成剧 | 一(引擎定义)、二(写成句子,尤其第 8 与第 10 条的二选一) | 三–七 |
| 设计人物网 | 三 | 五、六 |
| 选 pilot 类型 | 四 | 三、七 |
| 写 bible / pitch / series format | 六,字段表转 reference-documents.md | 一–四 |
| 诊断一份已有的方案或大纲 | 九(诊断清单);改单集时整组跳过 ㉑–㉚ | 六 |
| 改编 IP | 七 | 四、五 |
reference.md/reference-documents.md/reference-samples.md 都是查表用的,只在正文指向时打开。
Rabkin 给引擎原理,Landau 给字段,Oberg 给类型与规格,Douglas 给行业语汇,Goldberg/Rabkin 给工业底线,Blum 给文档层级,Miller/Calvisi/Smith 给表单,四部语料给证据。
1. franchise=concept + conflict + theme + story pattern(Rabkin ch5、ch8)。他自评 2011 年旧版的七项清单(人物、设定、故事类型、对白风格、互动方式、叙事风格、中心冲突)像用"灰皮长鼻大耳"描述大象:给现有剧写故事时是好 checklist,不能指导你造自己的引擎。
2. theme 禁止单词,必须是可争辩的对立命题:"There is no such thing as a one-word theme for a television series."(同章)The Shield——"一个好人在行善的追求中能做多少恶,才会变成他所对抗的恶?";Mad Men——"一个人以贩卖'消费带来幸福'为业,同时知道这在自己生活里不成立";Lost——"Faith versus reason"。
3. 主题必须落成人物之间的冲突("TV doesn't handle abstract concepts too well")。The Shield 的最小配置:能对抗恶的好人+一个恶+用传统方法对抗恶并失败的其他好人(否则观众不懂主角为何改用敌人的手段)。concept 只是容器、可替换(Lost 的 concept 来自 ABC 高管)。
4. 咬合测试(Rabkin ch6):"Quick, come up with a story that naturally explores both concept and conflict. Now come up with 99 more." 反例 Agents of SHIELD:唯一的中心冲突(Skye 既反集权又乐于加入 SHIELD)在一个人内部且与 concept 无关——"you can feel its emptiness";作者自己在 NBC 首季 Baywatch 同病(有 concept 无 conflict 无 theme,"every week you are reinventing the franchise")。
5. 类比不是小说,是爵士即兴(Rabkin ch4):冲突与主题=循环的和弦(chord changes),单集=在上面即兴的新旋律。
6. 同一件事的五种行业说法
7. story pattern=每集重复的那个形状(Rabkin ch8):Sons of Anarchy 不是"关于摩托帮",是 Jax 要保护三个家庭——察觉威胁→出手保护→反噬危及另一个家庭。→ 一句话写出 pattern,用 pilot 的后半段完整跑一遍(前半只做设定与触发)。(Breaking Bad 的 pattern 原文、结构常量与人物模式两层、反例见 reference.md 五)
8. 五道"有没有腿"的判据:99 个故事(Rabkin ch6)/六集测试(Landau 1e Tip 7)/pitch 备三、presentation 备五六(Landau 2e ch14/Blum ch17)/100 集与四问(Goldberg & Rabkin ch17)/五故事检验(Blum ch17)。(五项的阈值、原话与出处全表见 reference.md 七)
⚠️ 限定剧、迷你剧、一次性播完的闭合长篇不跑这一条。 100 集与"能撑五季"是美国广播网续订经济学的判据;一个封闭县域、一间办公室、一桩旧案这类题材按类型数本来就撑不到 100 集,照字面跑会把合格的项目否掉。闭合长篇改用第 10 条的引擎失压测算:算出物质性引擎在第几集失压,那一集就是这部剧该结束的地方,集数由此论证出来而不是被接受。两条判据二选一,按媒介与订购方式选。
9. 概念强弱不决定可持续性:Torchwood 反证(Rabkin ch9):按 concept 一定押 Torchwood,结果它只有四季而 Doctor Who 拍到第十季——外星威胁只是 premises/pretexts,不是 stories。→ "pilot 之后想不出三集"的病灶在人物冲突,不在 concept。(全案对照见 reference.md 五)
10. 引擎失压论(Succession S4 前言):判断一部剧该不该结束,看物质性情节引擎(交易、帝国、职位),不看人物还有没有戏——Armstrong 结束该剧不是因为人物写完了("the motor in the room was still running hot"),而是"the big plot engine, the brutal materialist plumbing which ran under the show, was looking to me like it would eventually start to lose pressure"。
用法:这一条是第 8 条的替代品,不是补充。可无限续订的剧用第 8 条(能不能再跑 100 集);闭合长篇与限定剧用这一条(引擎在第几集失压)。先判断媒介与订购方式,再选判据。
四个引擎样本各一句(详解见 reference.md 五):Sopranos=治疗室+两个家庭(Melfi 开场划下的"听到谋杀要报警"这条 ethical ground rule 本身是引擎);Fleabag=一张本身就在互相伤害的人物关系网+一个装置(对镜说话);坂元《花束般的恋爱》=双主角的 want 在五年间错位、全片没有反派;Succession="继承"这个永不能解决的问题。
11. 核心问题必须是过程句(Rabkin ch3):"Pilots need to raise questions that lead to processes, not solutions." 电影问"Who killed Marion Crane?";剧集问:Grantham 伯爵如何在不适合它的经济体系里保住庄园(Downton)?温和的化学老师如何在暴力的毒品生意里活下去(Breaking Bad)?Richie Finestra 能否在酒和毒品杀死他之前复兴摇滚(Vinyl)?→ 写成 "How will X keep doing Y in a world that Z?"。
12. 不要在 pilot 里把核心谜团演出来(同章,Extant 的三重错误是教科书案例):判决是 "execution is not the problem here. Conception is."(三重错误逐条见 reference.md 七)
13. Story landmines(Rabkin ch9)=pilot 潜台词里埋下、若干集后才引爆的人物冲突。→ 每个主要人物各埋一枚(未言明的恐惧/欲望/矛盾),写下预计第几集引爆;必须是人物层面——"她是警探的女儿"不算。(三条理由与 Doctor Who/Crazy Ex-Girlfriend 正例见 reference.md 七)
地雷数量与"一个总悬念"的冲突,以及解法。 八个主要人物=八枚地雷,在国产长剧里就等于八条并发谜线,而国产体系把"多谜并发"当病(见
sw-chinese-series-practice的"一个问题"格局与《皇城根儿》反例)。解法不是减少地雷,是改变它们的挂法:把全部地雷挂在同一个总悬念之下,当它的分解层,每集只揭一层,揭一层立刻补一个带时限的新悬念。于是观众追的始终是一个问题,而你仍然有八颗可定时引爆的炸药。排表时给每枚地雷写两个集号:引爆集,以及它揭开了总悬念的第几层。
14. 痛苦不是冲突;一个人的两个属性必须互相妨碍(同章,反例 Feed the Beast):"suffering isn't conflict, and so suffering alone can't drive story"。(Tommy/Dion 的逐条拆解与该剧 soap opera 化见 reference.md 七)
15. sweet spot(Landau 1e Tip 16)=类型、调性、主题、中心冲突、franchise、中心问题/谜团的交叉点;2e ch4 的公式是 sweet spot = character development × franchise,一小时剧的版本是 "a series must deliver the promise of its premise every week"——这就是 tacit contract。(原文、TWD/Homeland 范例与 currency 概念见 reference.md 七)
16. 营销实习生测试(Rabkin ch10)|可选,零成本复核,不要当一道工序——logline 写好之后再跑它,通常只是重读一遍并宣布通过;它的价值在"写 logline 的时候心里有这个读者",不在事后核对。:一句话能不能让一个初级营销人员替你向 Entertainment Weekly 要到几英寸版面。Bigger 不是预算大,是定义剧的冲突的规模。(对照实例见 reference.md 七)
17. "Big does not have to mean shallow." 三条加大法(同章):换职业(Sopranos 的起源只改了一处:职业)、换地点、换时代。反向立场(Miller ch1–2):"having great characters trumps high concept every time"。(三例全文与 Miller 的论据见 reference.md 七)
18. 记账式单元:每集"解决掉"的事,必须在后面某一集留下一笔要兑付的账。(本 pack 的推广,把 Succession S1 的季层做法下放到集层——原规则是"季终十条线全部落子,但每条落子都生成下一季的债务:Greg 的复印件、Tom 的邮轮、Roman 的过失致死、Shiv 的双重效忠",见 succession-series-writing。)
一般的案件剧到第 12 集开始重复,原因是每集结案等于清零。把"结案"改成"记账"就不会:本集签成的协议、保下的人、压住的投诉,都在账上留一个条目,由后面指定的某一集兑付。排大纲时给每条账目写成一行 第 N 集做了什么 → 第 M 集谁付账,M 必须是具体集号。一季里这样的账 6–10 笔为宜,过少则各集仍是孤岛,过多则观众记不住。
19. 用"同一套本事的第几种用法"给长剧分段,而不是用时间顺序。(本 pack 的推广,依据 Landau 1e Tip 13 的 movement 分季法:季初只定 tent poles,把季拆成每 4–6 集一个 movement,每个 movement 一件大事。)
主角的技能不变,用法换一次就是一个段落,换几次就有几段:学着用它赢 → 一边赢一边查/一边用它办私事 → 反过来用它并引爆更大的代价 → 放弃原目标、改用它去改规则。同一个人、四种用法、四种后果,天然就是四段;24 集除以四段=每段 6 集,正好落在 movement 的 4–6 集区间里。
20. 内在冲突必须外化成一圈拉扯他的人物(Rabkin ch6)。Buffy 的"想过正常生活 vs 拯救世界"外化为母亲要她正常、watcher 要她杀吸血鬼、好友在日常世界、爱人是百年吸血鬼;SHIELD 的 Skye 反之,冲突留在一个人内部,无法产生故事。
21. 每部剧都是家庭剧(Landau 2e ch6):"Every television series is about a family – whether they're related by blood or not." 推论:所有剧中家庭都必须 imperfect 且在某个层面 discontented。(六种家庭类型与例子见 reference.md 六)
22. familial role 分配库(同章):不必凑齐性别年龄,要确定"每个人物在正面和负面两个方向上会如何对待彼此"。(角色库全表与执法搭档/职场同事的推论见 reference.md 六)
23. transactional family(Logan Roy 范式)(同章):"寻求老板认可会取代寻求父母认可——而且前者还附带薪水作为额外确认";Logan "withholding his approval and validation","His relationships with his kids are all transactional."
24. 三条场景级规则(同章):"the people who love you most know how to hurt you the most."→写两人场前先确定"他知道对方哪个按钮",再决定按不按。(新人物测试与潜文本公式原文见 reference.md 六)
25. 专长分工(Landau 1e Tip 20):不同技能组合用来制造冲突、呈现多重 POV、帮观众区分人物;Grey's Anatomy 用专科定义人物。(Breaking Bad 分工表、House 的反同意原则、Out of Their Element 生成器见 reference.md 六)
26. 理性轴(Rabkin 附录 Mooresville pitch):"start from the position that they are both rational, intelligent adults" → 关系剧的双向吸引要给结构性理由,不只是化学反应。(全案见 reference.md 六)
27. 互补双主角(Landau 1e Tip 9):"two incomplete people together you get one whole."(Mulder/Scully=信与疑。)(Hacks 的 mutual need/shame 模板与 Gomorrah 的 logline 生成器见 reference.md 六)
28. 坂元式双主角错位(《花束般的恋爱》):不靠对抗而靠体贴造成价值转折(场 85 吹头发时说"我打算找工作了")。(四对镜像人物与台词预算分配见 reference.md 五)
29. 羞耻与空洞(Landau 2e ch7,引 Brené Brown《Daring Greatly》):shame = our fear of disconnection;创伤造成 emotional voids→voids 产生回避脆弱的 coping mechanisms。"Anyone who challenges any of these constructs becomes an antagonist."(三类防御的设计菜单见 reference.md 六)
30. grounded superpower 同时是 liability(Landau 2e ch5,Gilligan 的造人法):Walt 的超能力是"自欺的能力"。1e Tip 8:"when a character's greatest strength also becomes his or her greatest liability… The trick to dramatic sustainability is keeping your main character in this form of limbo."(三例与方框全文见 reference.md 六)
31. 同一 backstory 的两次陈述(Landau 2e ch7,Mare of Easttown):先给修饰版,几集后在更安全或更被迫的场合给真版本;落差就是人物的 shame。(两段台词原文与群像配平见 reference.md 六)
32. flaw 的揭示节奏(Landau 1e Tip 8):flaw 要源于 backstory 才能长跑,但 backstory 在 pilot 只斜向暗示——"what we don't know about a character is often more compelling than what we do know."(Don Draper/Scandal 的延迟揭示与 Girls 的反例见 reference.md 六)
33. 对抗力先定性质,再决定要不要一张脸(Landau 2e ch9):Succession=greed;Mare of Easttown 与 Fleabag=grief。总原则(脚注归 UCLA 荣休教授 Howard Suber):"All villains are antagonists, but not all antagonists are villains."(九部剧的性质对照与 Oberg 的补充见 reference.md 六)
34. 反派在自己的故事里是英雄(同章):"Every villain is the hero of his/her/their own story."——"they usually feel justified, entitled, and downright righteous"。结构判断:"as soon as that primary goal is successfully attained, the series is usually over – unless it pivots to another overarching mission or goal."(四条附则与 Rue 的错位案例见 reference.md 六)
35. 不要立草人(Kings 夫妇,Landau 2e ch4):"Propaganda is confirming what you already believe." 更值的是让我们喜欢的常规角色彼此不同意、且每个人都有道理。(全段与 The Good Wife 范例见 reference.md 六)
36. 可持续冲突网的三件套(Blum ch17 检验三):"如果所有连续剧主角都彼此相亲相爱,他们能做的就只能是微笑着彼此同意对方的意见了。" 引擎="团体/家庭纽带"的情节模式 + 每个角色预装的、与其他角色必然摩擦的性格属性 + 一个让他们无法离开彼此的巧合/处境;三者缺一撑不到三五年。(Downton 与 Succession 的逐项拆解、Succession 的三装置与 Tom 的反证见 reference.md 五;这张网填好的样本=Succession 七个 want/可失之物/绑定对象/引擎功能表,见 reference-samples.md 十三)
37. 三种分类(Landau 1e Tip 1/2e ch14;Douglas 3e ch2、Oberg 3.2、Calvisi §6 用不同名称说同一件事):premise pilot(第 1 集发生的事对"从第一天开始这部剧"是必要的)/non-premise/ongoing/midstream("It may be the audience's first episode, but it's not day one for the characters.")/hybrid(高潮时刻开场再 flash-forward 或 flashback);Oberg 另立 semi-cold start(The Walking Dead)。(四类定义表、例子清单、Sopranos 的归类与 47 部剧分类见 reference.md 四)
38. 两条进场硬规则:Oberg 3.2——premise pilot 不要把 inciting incident 留到集末,"You can't afford to have a pilot that's just a long first dramatic act.";Calvisi §6——"Enter the action of your story as late as possible in the world but as soon as possible in the script."(Scandal 原剧本里委托人第 12 页才进门,成片提前到第 6 分钟)。
39. 三种网台策略与"premise pilot 陷阱"(Blum ch17,1999–2001 语境,台别已过时、机制仍有效):原话是 premise pilot"很可能成为预测剧集发展走势的可怜晴雨表(a poor barometer)";Smith ch20 同向(半小时喜剧)——"most network executives prefer to see a typical-episode pilot";Landau 的不同立场:不是陷阱,只要求在 pitch 里说明是哪一种。(三种策略原文与 Smith 的论据见 reference.md 四) → 四条判据:①保守的大众频道或半小时喜剧→typical 或 half-premise;②世界设定复杂、必须解释规则→premise(Lost、Stranger Things、TWD);③人物关系网即引擎→midstream(Fleabag、The Office、Succession);④两头都要→Breaking Bad 式:前半设定、后半跑一遍 story pattern。优先级:四条可以同时命中(③人物网即引擎 + ②世界规则要解释是最常见的一对);命中两条以上一律走④——④是唯一能在一集里同时交付设定与"每周长什么样"的形状,而不是在②与③之间二选一。只命中一条时才按那一条办。
40. 三种收法:①**"The end of your pilot is the beginning of your series."(Landau);②Stick the landing**;③不一定要 cliffhanger(Miller ch6)。(三条原文与 Calvisi 的平衡句见 reference.md 四)
41. Douglas 的 pilot 四步(3e ch4):①Create the World(Lindelof:"the franchise is the world.")②Find the Story Springboards ③Populate the World ④Make a Plan。核心是秘密,深到要 88 或 100 小时才揭完;人物成长弧通常不是第一天的议程;前十页决定生死。(四步全文与引语见 reference.md 四)
42. 十个焦虑问题(Jeffries,USC 教 pilot 课):pilot 结束时列出至少十个让观众揪心的问题,这十个问题也用来推出十个未来集短梗概;先钩住高管,同时袖里留牌。(素材档案与那条永恒的 note 见 reference.md 四)
43. Sopranos 的"不解释机制"(Chase 前言 p0012+pilot):pilot 刻意不解释 Tony 的正式身份与"waste management"的运作,依据是 "rarely is anything black and white in life";这也是 pilot 不用 cliffhanger 收尾的理由。(全案与大台拒播的真实理由见 reference.md 五)
44. 八种 series type(Oberg 1.2):procedural(closed-ended)/serial/serial-procedural hybrid(三式)/sitcom/mini-series 与 limited/anthology/mini-series series/web series。sitcom 的不成文规则是 "characters don't change, and it's fine";limited(4–8 集)是从电影转剧集最容易的第一步。(八种定义、受众含义与 hybrid 三式全表见 reference.md 三)
45. mini-series series 的两条附加规则(Rabkin ch12):concept 不能 "defined by execution"(American Crime Story 的信息都在公共领域);anthology 与 mini-series series 的每季首集必须重新提供"连接观众与主角的即时叙事 hook"。Landau 2e ch1(2022):limited series 更难卖,新剧以 10 集/季为 benchmark。(全文与例外见 reference.md 三)
46. story-type 判定(Oberg 2.2):问 "Where does the main problem in the story come from, at season level?"——外部→plot-led;内部→character-led(不需要 antagonist 而需要 catalyst character);社会/story world→theme-led;另有 hybrid 与 exception。(三类的举例与例外机制见 reference.md 三)
47. 剧集专用的反向查法(同节):剧集几乎都多线,所以要反过来查——有没有一个主角和一个 main dramatic action/evolution 在季层成形?有则不是 theme-led。(第二个提示与 Chernobyl 的判定见 reference.md 三)
48. 选错类型就被锁死(Oberg 引言与 3.1 第 9 问):"The wrong combination of series type and story-type is the root cause of many series failures";最常见的 box-in 是character-led 剧在第一季把主角的内部问题完全解决(电影思维)→第二季必须换 story-type。(三例与平衡典范见 reference.md 三)
49. Maslow 与 M-Factor(Oberg 2.3、2.6):"the lower in the pyramid, the larger the potential audience",且必须垂直探索。M-Factor=受众、类型、story-type 相对 Maslow 的位置+预算是否自洽(1–5 分)。(Occupied/Money Heist/sitcom 的上下探索实例、whodunit 警告与 Breaking Bad 的动机植根见 reference.md 三)
50. What's at stake?(Oberg glossary)="开发人员不在乎故事时最常问的问题":主问题不解决会怎样,且必须让人悲伤/焦虑/恐惧;底线是不能有"说不出本集赌注"的集,且多线必须相连。(两条底线原文、Landau 的两条与半小时剧/关系剧的版本见 reference.md 七)
51. arena 的先天问题(Miller ch1,2015 年美国广播网视角):High school/Family/Relationship/News/Anthology 各有先天病灶,核心是 "A character's agency completes the story engine"——记者能"meddle"但无 license to intervene。(五类 arena 逐条、解法与适用范围的注记见 reference.md 三)
52. 三级文档(Blum ch17):创意 concept→连续剧陈述 series presentation→连续剧文献 series bible。(三级的原样定义与各自必含项见 reference.md 一)
53. 页数规格速查:bible 10–85 页/mini-bible 约 5 页;书面 pitch bible 10–30 页(甜点约 15 页);treatment drama 10–15 页/comedy 8–10 页;半小时喜剧 series format 10–15 页;口头 pitch 20 分钟 sweet spot,务必计时;mythology document +origin story 合计一页上限。(全表、出处与 Douglas 的 bible 三档见 reference.md 一;Oberg 的 bible 十二组件逐条规格与 bible/pitch deck/format 三者的区别见 reference-samples.md 十)
54. 四套字段清单(挑一套):Landau 2e ch1 Essential Guideposts(约 25 字段,也可当 bible 目录;Logline "Must contain the word 'but'";Central Question 关于未来/Central Mystery 关于过去)/Landau 2e ch14 的 11 项 pitch 配方/Miller 的 treatment 10 要素+六大问/Calvisi 的 17 栏 Basic Story Map(同一套栏目在"集"和"剧"两个层级各填一次)。(四套字段全文见 reference.md 一;Miller 的 10 要素在六份学生 treatment 里的字段对照矩阵见 reference-samples.md 十二)
55. 最容易写错的五处:①Overview 不是 pilot 摘要(Douglas 点名为"常见错误":要写整部剧的世界与 quest,并在 Overview 内就给出 springboards);②人物不要列清单;③不要把知道的全说出来(被追问才是成功);④不要过度规划;⑤文风必须与剧的类型一致。(五条的原话、各书补充与 Oberg/Landau/Smith 的硬规则见 reference.md 一)
56. 故事线文档骨架(本 pack 自拟,不属任何一本书)。各书都给了"整部剧"的文档,没有一本给"一条故事线"的文档;十字段表(线名/目的/推动者/赌注/本集入口/本集出口/跨集节拍/与主控思想的关系/与其他线的因果,含抽线测试)见 reference.md 二——用来在 break 季之前把每条线各写成一页再横向对齐。
Landau 的 story tentacles(2e ch4)是正面版:break story 时"choose the plotline that leads to more story"——找 fertile ground 而非 one-off dead ends。
(国产体系的改编五源类型学与逐集逐场改编实例见 sw-chinese-series-practice;舞台化细节见 sw-format-adaptation。以下除标注外出自 Landau 2e ch2。)
| 争点 | A | B | 用哪个 |
|---|---|---|---|
| 预先规划多季 | Rabkin ch4+Lindelof:"it's hubris to plot out… the second and third and fourth seasons"(St. Elsewhere 第五季把冲突都解决了,却续订第六季) | 同章引有线网高管:pitch 要"a strong, clear vision for the arc of the first season, and well-developed ideas for where subsequent years would go";Miller 要求 arc 三季 | pitch 按 B,进编剧室按 A;Oberg 折中:第一季清楚+后续季只写方向 |
| 进室前知不知道季终 | Armstrong:写第 1 集时已知道季终;Gomorrah 整季大纲先完成 | Gould(BCS):"feeling our way through it rather than having a master plan";Feldman 不知道 Steve 会在 S1 末死 | 群像多线剧倾向 A;靠演员化学反应演化的剧倾向 B |
| 先写剧本还是先 pitch | Landau:"write your pilot scripts on spec" | Aranya:"a script is also 60 pages of reasons to say no";Your Honor 只靠口头 pitch+三页纸卖掉 | 没履历按 A(Oberg 同);有履历或有包装按 B |
| premise 还是 typical pilot | Blum:"可怜的晴雨表";Smith:多数高管偏好 typical | Landau:不是陷阱,只要说明是哪一种 | 见第 39 条四判据 |
| 主题的地位 | Rabkin 把 theme 放四元素之首 | Kings 夫妇:"the number of times we actually used the word 'theme' in the writers' room is less than two" | 开发按 A(主题是 pitch 与过滤器),break story 可按 B |
| 人物会不会变 | Landau 2e:主角必须变;Calvisi:每季末要有 Assumption of Power | Oberg:"it's how the audience perceives them that changes";Armstrong 反对 "fallacy of growth"(人也会被 diminish);Shore:"nobody changes" | 可变的是处境与目标,核心常量由作者定义;限定剧可写完整弧 |
| 幕是不是结构 | Douglas:"Put away your books on three-act structure" | Oberg:四五幕只是 logistical format(香肠串比喻:只决定几根、多长) | 完整处理在 sw-series-structure;此处共识:引擎与幕数无关 |
| 要不要 high concept | Rabkin:不够大就没人抽样 | Miller:"great characters trumps high concept every time" | 并存:用 high concept 让高管肯读,用人物让剧活下去 |
| 家庭/关系剧引擎是否弱 | Miller(2015 广播网视角):两者都有先天问题 | Landau:每部剧都是家庭剧;Downton、Succession 的引擎来自阶级与继承 | Miller 的判断限于特定市场与年份;家庭群像要配强引擎或制度性摩擦 |
| arena 的真实性路线 | Gomorrah 实拍真街区(S1 一处酒吧外景一夜之间被炸掉;演保镖的当地人第五天失踪) | Costanzo 为安全在 Caserta 搭了整片那不勒斯 backlot | 取决于风险与预算;Blum:选址是"魅力 ÷ 可拍性"的商 |
引擎:①theme 是可争辩的对立命题而不是一个词?②主题落成两个人物之间的冲突、内部冲突外化了吗?③咬合测试能想出第 2 个、第 6 个故事吗?④核心问题是过程句还是答案句?⑤story pattern 能一句话写出、pilot 后半跑完一遍吗?⑥五六个后续故事的来源与结局形态雷同吗?⑦这组冲突能撑 100 集吗?(闭合长篇、限定剧、迷你剧跳过"能不能跑 100 集/五季"那一问,改跑第 10 条的引擎失压测算:算出物质性引擎在第几集失压,那一集就是该结束的地方。正文已这样分,此处同步。)⑧每集的赌注说得出来吗、在 Maslow 哪一层、各线有没有向下探?⑨主角是参与者还是观察者、有 license/agency 吗?⑩营销实习生测试:一句话推销只能说出"it's really good"吗?
人物网:⑪每个主要人物各带一枚潜台词地雷吗(人物层面而非情节揭秘),预计第几集引爆?⑫每人的两个属性互相干扰吗?⑬把痛苦当成冲突了吗、人物有可采取的行动吗?⑭对抗力的性质定了吗——定了之后才决定它要不要一张脸。⑮反派在自己的故事里是英雄吗?⑯每一对人物的关系向量写出来了吗(按哪个按钮、谁羡慕谁、谁看不惯谁)?⑰每个新人物都回答了"他的到来如何改变常规人物"吗?⑱主角的 grounded superpower 同时是 liability 吗?⑲有没有一处"同一 backstory 的两次陈述"?⑳主目标达成后剧是否就结束、有下一个 overarching mission 吗?
pilot 与文档(剧集层:只在立项、写 pilot、做提案时跑;改一个已播剧的单集请整组跳过):㉑pilot 是 premise/non-premise/hybrid 哪一种、为什么?㉒核心谜团被演出来了吗?㉓结尾留下十个揪心问题、能推出十个未来集吗?㉔pilot 演示了 franchise 吗(劫案已开始?治疗已开始?)㉕Overview 写成 pilot 摘要了吗、给了 springboards 吗?㉖人物是嵌在故事里介绍的还是一张清单?㉗文风与类型一致吗?㉘把知道的全说出来了吗(被追问才是成功)?㉙第五季写得太死吗?㉚character-led 剧第一季把内部问题解决干净了吗(最常见的 box-in)?
文档的读者:㉛这一页是给谁看的? 写作用文档(地雷表、按钮表、绑定物、故事线文档)与送审用文档(logline、overview、人物、pilot 梗概、后续故事)是两份东西,不要混在一份里交出去。平台审读不会看你的工具表;而把工具表删干净的"提案"又会让你自己在第 14 集断气。两份都写,只交一份。 判据:一页纸上如果出现了本 skill 的术语(franchise、landmine、对立统一、story pattern),它就是给你自己看的。
代价与分段:㉜每一集"解决掉"的事,有没有在后面某一集留下一笔要兑付的账?(见第二节 2.8)㉝全剧分段的依据是"主角用同一套本事的第几种用法",还是只是时间顺序?(见 2.9)
sw-series-structure)。sw-writers-room。© jtydhr88, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 3 other files in plugins/screenwriting/skills/sw-series-engine-bible of jtydhr88/screenwriting-skills.
Open the folder on GitHubat commit 51115f1
Sw Series Engine Bible next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Sw Series Engine Bible this skilljtydhr88/screenwriting-skills | 1.6k | — | ~6.7k | Automated safety check: Pass | MIT | |
| Idea To Blueprintsickn33/agentic-awesome-skills | 47k | 1 repos | ~6.1k | Automated safety check: Warn | MIT | |
| Story ReadinessDonchitos/Claude-Code-Game-Studios | 26k | — | ~6.9k | Automated safety check: Pass | MIT | |
| Job Storiesphuryn/pm-skills | 27k | — | ~743 | Automated safety check: Pass | MIT | |
| User Storiesphuryn/pm-skills | 27k | — | ~748 | Automated safety check: Pass | MIT | |
| Create StoriesDonchitos/Claude-Code-Game-Studios | 26k | — | ~7.3k | Automated safety check: Pass | MIT |
sickn33/agentic-awesome-skills
Turn a raw product, app, bot or feature idea into one evidence-backed build blueprint (researched stack, epics, Given/When/Then criteria, tests) that coding agents build one epic per session.
Donchitos/Claude-Code-Game-Studios
Is a story implementation-ready?. An agent skill from Donchitos/Claude-Code-Game-Studios.
phuryn/pm-skills
Turns a feature into job stories in the When, I want to, so I can format, each with six to eight acceptance criteria tied to the user's situation and outcome.
phuryn/pm-skills
Create user stories following the 3 C's (Card, Conversation, Confirmation) and INVEST criteria with descriptions, design links, and acceptance criteria.
Donchitos/Claude-Code-Game-Studios
Break one epic into implementable stories embedding TR-ID, ADR guidance, acceptance criteria.
sickn33/agentic-awesome-skills
Darwinian idea evolution engine — toss rough ideas onto an evolution island, let them compete, crossbreed, and mutate through structured rounds to surface your strongest concepts.
jtydhr88/screenwriting-skills
Writing screenplays in the style of Ozu Yasujiro and Noda Kogo (小津安二郎・野田高梧) — distilled from six complete screenplays (晚春 Late Spring, 麦秋 Early Summer, 东京物语 Tokyo Story, 浮草 Floating Weeds, 秋日和 Late…
jtydhr88/screenwriting-skills
Story structure for screenplays and stage plays (故事结构) — merged from Syd Field's paradigm and plot points, Blake Snyder's Save the Cat beat sheet and board, Robert McKee's Story…
jtydhr88/screenwriting-skills
Writing plays and screenplays in the Chekhovian mode (契诃夫戏剧法) — distilled from the complete plays (没有父亲的人/普拉东诺夫, 伊凡诺夫, 林妖→万尼亚舅舅, 海鸥, 三姊妹, 樱桃园, nine one-acts) plus Chekhov's own letters…
jtydhr88/screenwriting-skills
Writing a Succession-style ensemble series (继承之战群像剧写法) — distilled from Jesse Armstrong's complete shooting scripts for all four seasons, 39 episodes (Faber 2023), with his introductions and…
jtydhr88/screenwriting-skills
Case studies of American screenwriting (美国电影剧作案例) — Mei Feng's readings of the Hollywood Western (value pairs, four oppositions), screwball comedy (strong heroine, buddy vs adversary two-hander)…
jtydhr88/screenwriting-skills
Character creation, protagonist/antagonist design and dramatic conflict for scripts (人物/对手/冲突) — merged from Lajos Egri (three-dimensional character, orchestration, unity of opposites, pivotal…
From an idea to a sellable series design; series engine / franchise, story pattern, character webs that keep generating conflict, pilot types, and the documents that sell them…. Sw Series Engine Bible is an agent skill from jtydhr88/screenwriting-skills. From an idea to a sellable series design; series engine / franchise, story pattern, character webs that keep generating conflict, pilot types, and the documents that sell them (剧集引擎/franchise/故事模式/人物网/试播集类型/剧集圣经/跳板/提案).
Sw Series Engine Bible fits situations like: turning an idea into a series; testing whether a premise has legs; choosing premise vs typical-episode pilot; designing a sustainable character web.
Run `npx skills add jtydhr88/screenwriting-skills --skill sw-series-engine-bible -a claude-code`. Or copy the skill folder (plugins/screenwriting/skills/sw-series-engine-bible in jtydhr88/screenwriting-skills) into .claude/skills/sw-series-engine-bible in your project. Claude Code loads it when a task matches its description.
Run `npx skills add jtydhr88/screenwriting-skills --skill sw-series-engine-bible -a codex`. Or copy the skill folder (plugins/screenwriting/skills/sw-series-engine-bible in jtydhr88/screenwriting-skills) into .agents/skills/sw-series-engine-bible in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add jtydhr88/screenwriting-skills --skill sw-series-engine-bible -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sw-series-engine-bible, .gemini/skills/sw-series-engine-bible, .github/skills/sw-series-engine-bible and .opencode/skills/sw-series-engine-bible in your project.
SKILL.md names no scripts, command-line tools or credentials: Sw Series Engine Bible is instructions for the agent only.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Sw Series Engine Bible is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 6.7k tokens (SKILL.md is roughly 27k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Sw Series Engine Bible: Idea To Blueprint (sickn33/agentic-awesome-skills, 47k stars), Story Readiness (Donchitos/Claude-Code-Game-Studios, 26k stars), Job Stories (phuryn/pm-skills, 27k stars) and User Stories (phuryn/pm-skills, 27k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
jtydhr88 (a GitHub user) maintains it in jtydhr88/screenwriting-skills, which has 1,621 GitHub stars. The repository holds 25 skills in this directory. The repository was last updated on October 3, 2026.
Source: jtydhr88/screenwriting-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.