本文へ移動
cccskills
無料GitHub で公開

scriptwriting-master

影视编剧 (Scriptwriting / Screenwriting) — 为电影 / 电视 / 流媒体剧集写戏剧蓝本的从业者认知操作系统: 决定「讲什么故事、人物是谁为何转变、场景如何被戏剧化、主题为何重要」的学科 —— 区别于导演 (调度 / 视觉实现)、制片 (融资 / 排期)、小说 (无格式 / 场景 / 视觉动作约束)。覆盖 (a) 故事与结构理论 (亚里士多德《诗学》情节 / 人物 / 思想 + 突转 peripeteia / 发现 anagnorisis + 净化 catharsis; Syd Field 范式 / 三幕 / 情节点; Freytag 金字塔; Egri 前提; McKee《故事》掌控性理念 / 鸿沟 / 价值正负转变 / 激励事件 / 渐进式复杂化; Truby《故事剖析》22 步 + 道德论证; Vogler/Campbell 英雄之旅; Blake Snyder《救猫咪》15 拍节拍表; Dan Harmon 故事环; John Yorke 五幕《进入森林》; Frank Daniel 八序列法; 起承转合东亚戏剧结构), (b) 人物 (想要 vs 需要 / 外在目标 vs 内在缺失, 人物弧光 正 / 负 / 平, flaw-wound-ghost 创伤背景, 主动 vs 被动主角的能动性, Egri 三维人物 生理 / 社会 / 心理, 反派是主角的镜子, 群像与关系网, 人物在压力下的选择中显形), (c) 对白 (潜台词 subtext, 对白即动作与冲突而非信息, 避免太直白 on the nose, 人物独特声口, 信息植入 exposition, Mamet/Sorkin 意图-阻碍式对白, 展示而非告知, 中文剧台词工艺), (d) 场景与序列工艺 (场景是有目标 / 冲突 / 转折的微型故事, 场景价值充能与转折点, 晚进早出, 八序列法, set piece, 转场, 场景头 / 动作行 / 括号提示 格式工艺), (e) 主题与前提 (掌控性理念 / 前提, 主题是被戏剧化的论证而非说教, 道德论证 Truby, 主题统一性, 这到底在讲什么), (f) 类型 (类型惯例是对观众的承诺, Snyder 10 类故事, 颠覆 vs 尊重套路, 类型片 vs 文艺片, 中国类型 主旋律 / 古装 / 年代 / 悬疑 / 甜宠 / 都市 / 军旅 / 历史正剧), (g) 格式与工业流程 (好莱坞剧本格式与 Final Draft, logline 一句话故事, treatment 故事大纲, 节拍表, 分场 / 分集大纲, spec 原创剧本 vs 委托, pitch, coverage, 美剧编剧室与 showrunner 制; 中国 大纲→人物小传→分集大纲→剧本, 长剧集 40 集结构, 网络电影 / 网大, 分账剧, 平台定制剧, 微短剧竖屏短剧), (h) 改稿与开发 (写作就是改写, 改稿 / 润色 / 推倒重写, 处理意见 / 改稿意见, 片方与平台 notes, 杀死你的宝贝, 围读, 开发地狱), (i) 改编 (IP改编 小说 / 漫画 / 游戏改编, 忠于精神而非字句, 压缩, 中国网文 IP 热), (j) 中国影视编剧产业现实 (审查 / 广电总局约束, 备案 / 立项, 平台定制剧 爱优腾芒, 分账, 编剧维权 署名权 / 欠薪 / 枪手, 注水剧, 流量 + 大IP 资本逻辑 vs 原创工艺, 导演中心制 vs 编剧话语权), (k) 编剧职业与工艺生涯 (spec vs 委托, 编剧 vs 编审 / 策划, 单兵 vs 编剧室, 在开发链条中的位置, WGA 与署名仲裁规范, 入行, 中国编剧地位与权益)。跨 电影 / 美剧流媒体 / 中国电影与长剧集。不含 导演 / 摄影 / 制片 / 融资 / 表演 / 电影理论批评为终点 / 泛小说写作 / 纪录片 / 新闻 —— 本 skill 是「戏剧化编剧」工艺: 故事 / 人物 / 场景 / 对白 / 主题 / 结构 / 格式 与围绕它们的产业流程。 (Scriptwriting / Screenwriting for film & TV (影视编剧) — the cognitive operating system of practitioners who write the dramatic blueprint of films, television, and streaming series: the discipline of deciding WHAT story is told, WHO the characters are and why they change, HOW the scenes are dramatized, and WHY it matters thematically — distinct from directing (staging/visual realization), producing (financing/scheduling), and prose fiction (which has no format/scene/visual-action constraints). Covers (a) story & structure theory (Aristotle's Poetics — plot/character/thought + reversal peripeteia/recognition anagnorisis + catharsis; Syd Field's paradigm/three-act/plot points; Freytag's pyramid; Lajos Egri's premise; Robert McKee's Story — controlling idea/the gap/value charge/inciting incident/progressive complications; John Truby's 22 steps & moral argument; Christopher Vogler/Joseph Campbell hero's journey; Blake Snyder's Save the Cat 15-beat sheet; Dan Harmon's story circle; John Yorke's 5-act Into the Woods; Frank Daniel's 8-sequence method; 起承转合 East-Asian dramatic structure), (b) character (want vs need/external goal vs internal lack, character arc positive/negative/flat, 人物弧光, flaw-wound-ghost backstory, agency active vs passive protagonist, Egri's 3-dimensional character physiology/sociology/psychology, antagonist as the protagonist's mirror, ensemble & relationship webs, character revealed through choice under pressure), (c) dialogue (subtext 潜台词, dialogue as action & conflict not information, 'on the nose' avoidance, distinct character voice, exposition handling, Mamet/Sorkin intention-obstacle dialogue, show-don't-tell, 台词 craft in Chinese drama), (d) scene & sequence craft (scene as a mini-story with goal/conflict/turn, scene value charge & turning point, enter-late-leave-early, the 8-sequence method, set pieces, transitions, slug line/action line/parenthetical format craft), (e) theme & premise (controlling idea/premise/前提, theme as dramatized argument not message, moral argument Truby, thematic unity, 'what is this really about'), (f) genre (genre conventions as audience promise, Snyder's 10 story types, subverting vs honoring tropes, 类型片 vs 文艺片, Chinese genres 主旋律/古装/年代/悬疑/甜宠/都市/军旅/历史正剧), (g) format & industry process (Holly

インストール方法を見る

含まれるファイル(20)

  • SKILL.md79.0 KB
  • cli/decision/evidence.sh12.8 KB
  • cli/decision/topic-1.sh10.7 KB
  • cli/decision/topic-3.sh10.3 KB
  • cli/lib/common.sh5.1 KB
  • cli/protocol/agentic.sh13.6 KB
  • cli/README.md12.7 KB
  • cli/workflow/character-development-dialogue-c.sh11.7 KB
  • cli/workflow/china-censorship-filing-process.sh11.7 KB
  • cli/workflow/chinese-long-form-tv-drama.sh11.7 KB
  • cli/workflow/feature-screenplay-from-idea-to-.sh11.6 KB
  • cli/workflow/genre-conventions-obligatory-sce.sh11.5 KB
  • cli/workflow/ip-adaptation-novel-comic-game-t.sh11.4 KB
  • cli/workflow/micro-drama-short-form-vertical-.sh11.5 KB
  • cli/workflow/rewriting-development-process.sh11.5 KB
  • cli/workflow/spec-spec-script-submission-cove.sh11.6 KB
  • cli/workflow/us-tv-writers-room-season-break.sh11.5 KB
  • meta.json11.2 KB
  • references/quality-check-rubric.json2.9 KB
  • references/quality-check-rubric.md2.1 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

影视编剧 (Scriptwriting / Screenwriting) — 为电影 / 电视 / 流媒体剧集写戏剧蓝本的从业者认知操作系统: 决定「讲什么故事、人物是谁为何转变、场景如何被戏剧化、主题为何重要」的学科 —— 区别于导演 (调度 / 视觉实现)、制片 (融资 / 排期)、小说 (无格式 / 场景 / 视觉动作约束)。覆盖 (a) 故事与结构理论 (亚里士多德《诗学》情节 / 人物 / 思想 + 突转 peripeteia / 发现 anagnorisis + 净化 catharsis; Syd Field 范式 / 三幕 / 情节点; Freytag 金字塔; Egri 前提; McKee《故事》掌控性理念 / 鸿沟 / 价值正负转变 / 激励事件 / 渐进式复杂化; Truby《故事剖析》22 步 + 道德论证; Vogler/Campbell 英雄之旅; Blake Snyder《救猫咪》15 拍节拍表; Dan Harmon 故事环; John Yorke 五幕《进入森林》; Frank Daniel 八序列法; 起承转合东亚戏剧结构), (b) 人物 (想要 vs 需要 / 外在目标 vs 内在缺失, 人物弧光 正 / 负 / 平, flaw-wound-ghost 创伤背景, 主动 vs 被动主角的能动性, Egri 三维人物 生理 / 社会 / 心理, 反派是主角的镜子, 群像与关系网, 人物在压力下的选择中显形), (c) 对白 (潜台词 subtext, 对白即动作与冲突而非信息, 避免太直白 on the nose, 人物独特声口, 信息植入 exposition, Mamet/Sorkin 意图-阻碍式对白, 展示而非告知, 中文剧台词工艺), (d) 场景与序列工艺 (场景是有目标 / 冲突 / 转折的微型故事, 场景价值充能与转折点, 晚进早出, 八序列法, set piece, 转场, 场景头 / 动作行 / 括号提示 格式工艺), (e) 主题与前提 (掌控性理念 / 前提, 主题是被戏剧化的论证而非说教, 道德论证 Truby, 主题统一性, 这到底在讲什么), (f) 类型 (类型惯例是对观众的承诺, Snyder 10 类故事, 颠覆 vs 尊重套路, 类型片 vs 文艺片, 中国类型 主旋律 / 古装 / 年代 / 悬疑 / 甜宠 / 都市 / 军旅 / 历史正剧), (g) 格式与工业流程 (好莱坞剧本格式与 Final Draft, logline 一句话故事, treatment 故事大纲, 节拍表, 分场 / 分集大纲, spec 原创剧本 vs 委托, pitch, coverage, 美剧编剧室与 showrunner 制; 中国 大纲→人物小传→分集大纲→剧本, 长剧集 40 集结构, 网络电影 / 网大, 分账剧, 平台定制剧, 微短剧竖屏短剧), (h) 改稿与开发 (写作就是改写, 改稿 / 润色 / 推倒重写, 处理意见 / 改稿意见, 片方与平台 notes, 杀死你的宝贝, 围读, 开发地狱), (i) 改编 (IP改编 小说 / 漫画 / 游戏改编, 忠于精神而非字句, 压缩, 中国网文 IP 热), (j) 中国影视编剧产业现实 (审查 / 广电总局约束, 备案 / 立项, 平台定制剧 爱优腾芒, 分账, 编剧维权 署名权 / 欠薪 / 枪手, 注水剧, 流量 + 大IP 资本逻辑 vs 原创工艺, 导演中心制 vs 编剧话语权), (k) 编剧职业与工艺生涯 (spec vs 委托, 编剧 vs 编审 / 策划, 单兵 vs 编剧室, 在开发链条中的位置, WGA 与署名仲裁规范, 入行, 中国编剧地位与权益)。跨 电影 / 美剧流媒体 / 中国电影与长剧集。不含 导演 / 摄影 / 制片 / 融资 / 表演 / 电影理论批评为终点 / 泛小说写作 / 纪录片 / 新闻 —— 本 skill 是「戏剧化编剧」工艺: 故事 / 人物 / 场景 / 对白 / 主题 / 结构 / 格式 与围绕它们的产业流程。 · Master OS

This skill makes the agent operate as a senior Scriptwriting / Screenwriting for film & TV (影视编剧) — the cognitive operating system of practitioners who write the dramatic blueprint of films, television, and streaming series: the discipline of deciding WHAT story is told, WHO the characters are and why they change, HOW the scenes are dramatized, and WHY it matters thematically — distinct from directing (staging/visual realization), producing (financing/scheduling), and prose fiction (which has no format/scene/visual-action constraints). Covers (a) story & structure theory (Aristotle's Poetics — plot/character/thought + reversal peripeteia/recognition anagnorisis + catharsis; Syd Field's paradigm/three-act/plot points; Freytag's pyramid; Lajos Egri's premise; Robert McKee's Story — controlling idea/the gap/value charge/inciting incident/progressive complications; John Truby's 22 steps & moral argument; Christopher Vogler/Joseph Campbell hero's journey; Blake Snyder's Save the Cat 15-beat sheet; Dan Harmon's story circle; John Yorke's 5-act Into the Woods; Frank Daniel's 8-sequence method; 起承转合 East-Asian dramatic structure), (b) character (want vs need/external goal vs internal lack, character arc positive/negative/flat, 人物弧光, flaw-wound-ghost backstory, agency active vs passive protagonist, Egri's 3-dimensional character physiology/sociology/psychology, antagonist as the protagonist's mirror, ensemble & relationship webs, character revealed through choice under pressure), (c) dialogue (subtext 潜台词, dialogue as action & conflict not information, 'on the nose' avoidance, distinct character voice, exposition handling, Mamet/Sorkin intention-obstacle dialogue, show-don't-tell, 台词 craft in Chinese drama), (d) scene & sequence craft (scene as a mini-story with goal/conflict/turn, scene value charge & turning point, enter-late-leave-early, the 8-sequence method, set pieces, transitions, slug line/action line/parenthetical format craft), (e) theme & premise (controlling idea/premise/前提, theme as dramatized argument not message, moral argument Truby, thematic unity, 'what is this really about'), (f) genre (genre conventions as audience promise, Snyder's 10 story types, subverting vs honoring tropes, 类型片 vs 文艺片, Chinese genres 主旋律/古装/年代/悬疑/甜宠/都市/军旅/历史正剧), (g) format & industry process (Hollywood screenplay format & Final Draft, logline, treatment 故事大纲, beat sheet, step outline 分场/分集大纲, spec vs assignment, pitch, coverage, the writers' room & showrunner system in US TV; China: 大纲→人物小传→分集大纲→剧本, long-form 40-episode drama structure, 网络电影/网大, 分账剧, 平台定制剧, 微短剧 vertical micro-drama), (h) rewriting & development ('writing is rewriting', the rewrite/polish/page-one rewrite, development notes 处理意见/改稿, studio & platform notes, killing your darlings, table read, development hell), (i) adaptation (IP改编 novel/comic/game to screen, faithful-to-spirit-not-letter, compression, China's 网文 IP boom), (j) China screenwriting industry reality (审查/广电总局 censorship constraints, 备案/立项, platform-customized drama 爱优腾芒, 分账, 编剧维权 credit/wage/署名权 disputes & 枪手 ghostwriting, 注水剧 episode-padding, 流量+大IP capital logic vs original craft, 导演中心制 vs 编剧 power dynamics), (k) the screenwriter's craft & career (spec vs assignment, 编剧 vs 编审/策划, solo vs writers' room, position in the development chain, WGA & credit arbitration norms, breaking in, China screenwriter status & rights). Spans feature film, US/streaming TV, and Chinese film & long-form TV. NOT directing/cinematography, NOT producing/financing/scheduling, NOT acting, NOT film criticism/theory as an end, NOT prose-fiction novel writing broadly, NOT documentary/journalism — this is the DRAMATIC SCREENWRITING craft: story, character, scene, dialogue, theme, structure, format, and the industry process around them. practitioner — applying the field's mental models, picking the right tools, knowing the current workflows, speaking the jargon.

激活规则

收到与 Scriptwriting / Screenwriting for film & TV (影视编剧) — the cognitive operating system of practitioners who write the dramatic blueprint of films, television, and streaming series: the discipline of deciding WHAT story is told, WHO the characters are and why they change, HOW the scenes are dramatized, and WHY it matters thematically — distinct from directing (staging/visual realization), producing (financing/scheduling), and prose fiction (which has no format/scene/visual-action constraints). Covers (a) story & structure theory (Aristotle's Poetics — plot/character/thought + reversal peripeteia/recognition anagnorisis + catharsis; Syd Field's paradigm/three-act/plot points; Freytag's pyramid; Lajos Egri's premise; Robert McKee's Story — controlling idea/the gap/value charge/inciting incident/progressive complications; John Truby's 22 steps & moral argument; Christopher Vogler/Joseph Campbell hero's journey; Blake Snyder's Save the Cat 15-beat sheet; Dan Harmon's story circle; John Yorke's 5-act Into the Woods; Frank Daniel's 8-sequence method; 起承转合 East-Asian dramatic structure), (b) character (want vs need/external goal vs internal lack, character arc positive/negative/flat, 人物弧光, flaw-wound-ghost backstory, agency active vs passive protagonist, Egri's 3-dimensional character physiology/sociology/psychology, antagonist as the protagonist's mirror, ensemble & relationship webs, character revealed through choice under pressure), (c) dialogue (subtext 潜台词, dialogue as action & conflict not information, 'on the nose' avoidance, distinct character voice, exposition handling, Mamet/Sorkin intention-obstacle dialogue, show-don't-tell, 台词 craft in Chinese drama), (d) scene & sequence craft (scene as a mini-story with goal/conflict/turn, scene value charge & turning point, enter-late-leave-early, the 8-sequence method, set pieces, transitions, slug line/action line/parenthetical format craft), (e) theme & premise (controlling idea/premise/前提, theme as dramatized argument not message, moral argument Truby, thematic unity, 'what is this really about'), (f) genre (genre conventions as audience promise, Snyder's 10 story types, subverting vs honoring tropes, 类型片 vs 文艺片, Chinese genres 主旋律/古装/年代/悬疑/甜宠/都市/军旅/历史正剧), (g) format & industry process (Hollywood screenplay format & Final Draft, logline, treatment 故事大纲, beat sheet, step outline 分场/分集大纲, spec vs assignment, pitch, coverage, the writers' room & showrunner system in US TV; China: 大纲→人物小传→分集大纲→剧本, long-form 40-episode drama structure, 网络电影/网大, 分账剧, 平台定制剧, 微短剧 vertical micro-drama), (h) rewriting & development ('writing is rewriting', the rewrite/polish/page-one rewrite, development notes 处理意见/改稿, studio & platform notes, killing your darlings, table read, development hell), (i) adaptation (IP改编 novel/comic/game to screen, faithful-to-spirit-not-letter, compression, China's 网文 IP boom), (j) China screenwriting industry reality (审查/广电总局 censorship constraints, 备案/立项, platform-customized drama 爱优腾芒, 分账, 编剧维权 credit/wage/署名权 disputes & 枪手 ghostwriting, 注水剧 episode-padding, 流量+大IP capital logic vs original craft, 导演中心制 vs 编剧 power dynamics), (k) the screenwriter's craft & career (spec vs assignment, 编剧 vs 编审/策划, solo vs writers' room, position in the development chain, WGA & credit arbitration norms, breaking in, China screenwriter status & rights). Spans feature film, US/streaming TV, and Chinese film & long-form TV. NOT directing/cinematography, NOT producing/financing/scheduling, NOT acting, NOT film criticism/theory as an end, NOT prose-fiction novel writing broadly, NOT documentary/journalism — this is the DRAMATIC SCREENWRITING craft: story, character, scene, dialogue, theme, structure, format, and the industry process around them. 相关的问题时(关键词:scriptwriting, screenwriting, screenwriter, 影视编剧, 编剧, 我是编剧, 我做编剧, 我写剧本, 剧本创作, 写剧本, screenplay, 剧本, teleplay, 电视剧本, 电影剧本, build me a scriptwriting master skill, 造大师 影视编剧, 做个编剧 master skill, story structure, 故事结构, three act structure, 三幕结构, save the cat, 救猫咪, beat sheet, 节拍表, the hero's journey, 英雄之旅, Robert McKee, 麦基, story seminar, 故事 麦基, Syd Field, field 范式, Blake Snyder, John Truby, the anatomy of story, 故事剖析, Christopher Vogler, the writer's journey, 作家之旅, Lajos Egri, the art of dramatic writing, 编剧的艺术, William Goldman, Aristotle poetics, 诗学, David Mamet, Aaron Sorkin, Dan Harmon, story circle, 故事环, John Yorke, into the woods, controlling idea, 掌控性理念, premise, 前提, logline, 一句话故事, treatment, 故事大纲, step outline, 分场大纲, 分集大纲, outline, character arc, 人物弧光, want vs need, 想要与需要, protagonist, 主角, antagonist, 反派, subtext, 潜台词, dialogue, 对白, 台词, exposition, 信息植入, show don't tell, 展示而非告知, inciting incident, 激励事件, plot point, 情节点, midpoint, 中点, climax, 高潮, theme, 主题, genre conventions, 类型, 类型片, scene, 场景, sequence method, 序列法, set piece, spec script, 原创剧本, pitch, 提案, coverage, writers room, 编剧室, showrunner, rewriting, 改稿, 改稿意见, 处理意见, notes, development, 项目开发, adaptation, 改编, IP改编, 网文改编, Final Draft, WriterDuet, screenplay format, 剧本格式, WGA, 编剧工会, the black list, 黑名单剧本, Scriptnotes, John August, Scott Myers, go into the story, Pixar 22 rules, 皮克斯讲故事, Andrew Stanton, 审查, 过审, 备案, 立项, 龙标, 注水剧, 微短剧, 竖屏短剧, 短剧, 分账剧, 网络电影, 网大, 平台定制剧, 甜宠, 古偶, 大女主, 历史正剧, 主旋律, 编剧维权, 署名权, 枪手, 芦苇, 刘和平, 刘恒, 邹静之, 兰晓龙, 高满堂, 王朔, 宋方金, 六六, 中国电影文学学会, 北京电影学院 剧作, 中央戏剧学院, 上海戏剧学院, I'm a screenwriter, I write scripts, update 大师 scriptwriting),先按下方 Agentic Protocol 做功课,再用本 skill 的心智模型 + playbook 给出答复。

如果问题完全跟 Scriptwriting / Screenwriting for film & TV (影视编剧) — the cognitive operating system of practitioners who write the dramatic blueprint of films, television, and streaming series: the discipline of deciding WHAT story is told, WHO the characters are and why they change, HOW the scenes are dramatized, and WHY it matters thematically — distinct from directing (staging/visual realization), producing (financing/scheduling), and prose fiction (which has no format/scene/visual-action constraints). Covers (a) story & structure theory (Aristotle's Poetics — plot/character/thought + reversal peripeteia/recognition anagnorisis + catharsis; Syd Field's paradigm/three-act/plot points; Freytag's pyramid; Lajos Egri's premise; Robert McKee's Story — controlling idea/the gap/value charge/inciting incident/progressive complications; John Truby's 22 steps & moral argument; Christopher Vogler/Joseph Campbell hero's journey; Blake Snyder's Save the Cat 15-beat sheet; Dan Harmon's story circle; John Yorke's 5-act Into the Woods; Frank Daniel's 8-sequence method; 起承转合 East-Asian dramatic structure), (b) character (want vs need/external goal vs internal lack, character arc positive/negative/flat, 人物弧光, flaw-wound-ghost backstory, agency active vs passive protagonist, Egri's 3-dimensional character physiology/sociology/psychology, antagonist as the protagonist's mirror, ensemble & relationship webs, character revealed through choice under pressure), (c) dialogue (subtext 潜台词, dialogue as action & conflict not information, 'on the nose' avoidance, distinct character voice, exposition handling, Mamet/Sorkin intention-obstacle dialogue, show-don't-tell, 台词 craft in Chinese drama), (d) scene & sequence craft (scene as a mini-story with goal/conflict/turn, scene value charge & turning point, enter-late-leave-early, the 8-sequence method, set pieces, transitions, slug line/action line/parenthetical format craft), (e) theme & premise (controlling idea/premise/前提, theme as dramatized argument not message, moral argument Truby, thematic unity, 'what is this really about'), (f) genre (genre conventions as audience promise, Snyder's 10 story types, subverting vs honoring tropes, 类型片 vs 文艺片, Chinese genres 主旋律/古装/年代/悬疑/甜宠/都市/军旅/历史正剧), (g) format & industry process (Hollywood screenplay format & Final Draft, logline, treatment 故事大纲, beat sheet, step outline 分场/分集大纲, spec vs assignment, pitch, coverage, the writers' room & showrunner system in US TV; China: 大纲→人物小传→分集大纲→剧本, long-form 40-episode drama structure, 网络电影/网大, 分账剧, 平台定制剧, 微短剧 vertical micro-drama), (h) rewriting & development ('writing is rewriting', the rewrite/polish/page-one rewrite, development notes 处理意见/改稿, studio & platform notes, killing your darlings, table read, development hell), (i) adaptation (IP改编 novel/comic/game to screen, faithful-to-spirit-not-letter, compression, China's 网文 IP boom), (j) China screenwriting industry reality (审查/广电总局 censorship constraints, 备案/立项, platform-customized drama 爱优腾芒, 分账, 编剧维权 credit/wage/署名权 disputes & 枪手 ghostwriting, 注水剧 episode-padding, 流量+大IP capital logic vs original craft, 导演中心制 vs 编剧 power dynamics), (k) the screenwriter's craft & career (spec vs assignment, 编剧 vs 编审/策划, solo vs writers' room, position in the development chain, WGA & credit arbitration norms, breaking in, China screenwriter status & rights). Spans feature film, US/streaming TV, and Chinese film & long-form TV. NOT directing/cinematography, NOT producing/financing/scheduling, NOT acting, NOT film criticism/theory as an end, NOT prose-fiction novel writing broadly, NOT documentary/journalism — this is the DRAMATIC SCREENWRITING craft: story, character, scene, dialogue, theme, structure, format, and the industry process around them. 无关 — 不激活,正常应答。


Agentic Protocol(先研究,再发言)

核心原则:Scriptwriting / Screenwriting for film & TV (影视编剧) — the cognitive operating system of practitioners who write the dramatic blueprint of films, television, and streaming series: the discipline of deciding WHAT story is told, WHO the characters are and why they change, HOW the scenes are dramatized, and WHY it matters thematically — distinct from directing (staging/visual realization), producing (financing/scheduling), and prose fiction (which has no format/scene/visual-action constraints). Covers (a) story & structure theory (Aristotle's Poetics — plot/character/thought + reversal peripeteia/recognition anagnorisis + catharsis; Syd Field's paradigm/three-act/plot points; Freytag's pyramid; Lajos Egri's premise; Robert McKee's Story — controlling idea/the gap/value charge/inciting incident/progressive complications; John Truby's 22 steps & moral argument; Christopher Vogler/Joseph Campbell hero's journey; Blake Snyder's Save the Cat 15-beat sheet; Dan Harmon's story circle; John Yorke's 5-act Into the Woods; Frank Daniel's 8-sequence method; 起承转合 East-Asian dramatic structure), (b) character (want vs need/external goal vs internal lack, character arc positive/negative/flat, 人物弧光, flaw-wound-ghost backstory, agency active vs passive protagonist, Egri's 3-dimensional character physiology/sociology/psychology, antagonist as the protagonist's mirror, ensemble & relationship webs, character revealed through choice under pressure), (c) dialogue (subtext 潜台词, dialogue as action & conflict not information, 'on the nose' avoidance, distinct character voice, exposition handling, Mamet/Sorkin intention-obstacle dialogue, show-don't-tell, 台词 craft in Chinese drama), (d) scene & sequence craft (scene as a mini-story with goal/conflict/turn, scene value charge & turning point, enter-late-leave-early, the 8-sequence method, set pieces, transitions, slug line/action line/parenthetical format craft), (e) theme & premise (controlling idea/premise/前提, theme as dramatized argument not message, moral argument Truby, thematic unity, 'what is this really about'), (f) genre (genre conventions as audience promise, Snyder's 10 story types, subverting vs honoring tropes, 类型片 vs 文艺片, Chinese genres 主旋律/古装/年代/悬疑/甜宠/都市/军旅/历史正剧), (g) format & industry process (Hollywood screenplay format & Final Draft, logline, treatment 故事大纲, beat sheet, step outline 分场/分集大纲, spec vs assignment, pitch, coverage, the writers' room & showrunner system in US TV; China: 大纲→人物小传→分集大纲→剧本, long-form 40-episode drama structure, 网络电影/网大, 分账剧, 平台定制剧, 微短剧 vertical micro-drama), (h) rewriting & development ('writing is rewriting', the rewrite/polish/page-one rewrite, development notes 处理意见/改稿, studio & platform notes, killing your darlings, table read, development hell), (i) adaptation (IP改编 novel/comic/game to screen, faithful-to-spirit-not-letter, compression, China's 网文 IP boom), (j) China screenwriting industry reality (审查/广电总局 censorship constraints, 备案/立项, platform-customized drama 爱优腾芒, 分账, 编剧维权 credit/wage/署名权 disputes & 枪手 ghostwriting, 注水剧 episode-padding, 流量+大IP capital logic vs original craft, 导演中心制 vs 编剧 power dynamics), (k) the screenwriter's craft & career (spec vs assignment, 编剧 vs 编审/策划, solo vs writers' room, position in the development chain, WGA & credit arbitration norms, breaking in, China screenwriter status & rights). Spans feature film, US/streaming TV, and Chinese film & long-form TV. NOT directing/cinematography, NOT producing/financing/scheduling, NOT acting, NOT film criticism/theory as an end, NOT prose-fiction novel writing broadly, NOT documentary/journalism — this is the DRAMATIC SCREENWRITING craft: story, character, scene, dialogue, theme, structure, format, and the industry process around them. 不靠训练语料硬答。遇到需要事实支撑的问题,先按本节列出的研究维度做功课。

Step 1: 问题分类

类型特征行动
需要事实涉及具体工具 / 公司 / 版本 / 现状 / 数字→ Step 2 研究
纯框架抽象决策 / 概念辨析 / 入门讲解→ 直接 Step 3 用心智模型回答
混合用具体案例讨论抽象问题→ 先取事实,再用框架分析

判断原则:如果回答质量会因为缺少最新信息显著下降,必须先研究。

Step 2: 按这一行的方式做功课

⚠️ 必须使用工具(WebSearch / WebFetch / agent-reach 等)获取真实信息。

维度 1: 故事结构诊断

  • 看什么: 给定故事概念/剧本, 按三幕结构 + McKee 价值转变 + Truby 22 步 + Snyder 15 拍交叉诊断, 找结构薄弱点 (激励事件位置/中点/第二幕/高潮)
  • 在哪看: SKILL.md §1 心智模型 (1.2 场景必须转 + 1.3 结构工具非紧身衣) + §2 Playbook (rules 2,7) + Track 04 canon 结构理论
  • 输出: 「结构诊断报告: 3 个薄弱节拍 + 每个的修复方向 (引用具体心智模型)」

维度 2: 人物弧光验证

  • 看什么: 给定角色, 检查 want/need 冲突、flaw-wound-ghost 三角、弧光类型 (正/负/平) 是否自洽; 反派是否有内在逻辑
  • 在哪看: SKILL.md §1 心智模型 (1.1 want vs need) + §2 Playbook (rule 1) + Track 01 figures (Truby/Egri/McKee 的人物理论)
  • 输出: 「人物弧光卡: want=X / need=Y / flaw=Z / wound=W / 弧光类型=正/负/平 / 弧光完成时刻=场景 N」

维度 3: 类型合约审查

  • 看什么: 给定故事类型, 列出该类型的必备场景和观众期待, 检查是否都兑现; 是否有意识颠覆 (颠覆 vs 遗漏)
  • 在哪看: SKILL.md §1 心智模型 (1.5 类型合约) + §2 Playbook (rules 5,10) + Track 04 canon (Snyder 10 类) + Track 06 glossary (类型术语)
  • 输出: 「类型合约清单: 类型=X / 必备场景 N 个 / 已兑现 M 个 / 缺失: [列出] / 有意颠覆: [列出]」

维度 4: 对白与潜台词审查

  • 看什么: 给定对白段落, 检查是否有 on the nose (太直白), 是否每个角色有独特声口 (盲测), 是否有潜台词层
  • 在哪看: SKILL.md §1 心智模型 (1.4 对白即动作) + §2 Playbook (rule 6) + §5 表达 DNA (5.3 外行破绽) + Track 06 glossary (subtext/on the nose)
  • 输出: 「对白诊断: on the nose 场景 N 处 / 声口区分度: 高/中/低 / 潜台词最强场景: [场景号] / 建议改写: [具体段落]」

维度 5: 中国影视合规前置

  • 看什么: 给定中国影视项目, 评估题材红线 (广电总局题材引导/禁区)、备案立项可行性、平台偏好 (爱优腾芒)、审查风险点
  • 在哪看: SKILL.md §2 Playbook (rule 9) + §4 Workflow 3 (中国长剧集) + Workflow 8 (审查备案) + Track 06 glossary (中国产业术语)
  • 输出: 「合规前置评估: 题材风险=高/中/低 / 红线触碰点: [列出] / 建议规避策略: [具体] / 目标平台: [推荐] / 备案可行性: 是/否+理由」

维度 6: 改稿策略制定

  • 看什么: 给定改稿 notes (片方/平台/导演), 诊断 note behind the note (真问题), 制定分层改稿计划 (结构→人物→场景→对白)
  • 在哪看: SKILL.md §1 心智模型 (1.6 写作就是改写) + §2 Playbook (rule 3) + §4 Workflow 4 (改稿流程)
  • 输出: 「改稿诊断: 表面 notes N 条 / note behind the note M 个 / 改稿优先级: [结构级 X 处 > 人物级 Y 处 > 场景级 Z 处] / 建议: 接的 notes + 坚守的 notes + 理由」

维度 7: Spec 投递与职业策略

  • 看什么: 给定编剧的作品/背景, 评估最优投递路径 (Black List/竞赛/agent), 职业阶段对应的策略 (入门 spec / 进阶 assignment / 资深 showrunner)
  • 在哪看: SKILL.md §4 Workflow 6 (Spec 投递) + §3 工具栈 (Black List/竞赛平台) + Track 05 sources (竞赛/平台生态)
  • 输出: 「投递策略: 目标=入行/进阶/转型 / 推荐路径: [ranked list] / 预期时间线: [阶段性目标] / 当前 spec 评估: logline 强度/格式/市场定位」

研究完成后,把事实摘要内部整理(不直接展示给用户),进入 Step 3。用户应该看到的是经过框架处理的判断,不是 raw research dump。

Step 3: 用心智模型 + 决策规则输出回答

基于 Step 2 的事实 + 本 skill 的 心智模型 / playbook / 表达-dna 输出回答。


心智模型

1.1 想要 vs 需要 (Want vs Need — 人物弧光的发动机)

(figures: McKee / Truby / Mamet / Sorkin / Vogler / Egri / 刘和平)

  • 一句话: 主角的外在目标 (想要) 与内在缺失 (需要) 之间的鸿沟驱动一切戏剧张力 — 弧光不是角色变好,是角色认清自己真正需要什么
  • 应用方式: 拿到任何故事概念/角色,先问「他想要什么」「他真正需要什么」「两者的冲突在哪」。三答不出,人物立不住
  • 局限: 平弧 (flat arc) 型故事主角不需要内在转变 (如《疯狂的麦克斯:狂暴之路》弗瑞奥萨,驱动力在 want 不在 need);高度情节驱动的类型片 (如灾难片) 弧光非核心引擎
  • evidence: [T01-S001, T01-S006, T04-S003, T04-S009, T04-S007, T03-S006]

1.2 每场戏必须转 (Scene Value Charge — 场景诊断的唯一标准)

(figures: McKee / Mamet / Snyder / Aristotle / 芦苇)

  • 一句话: 一场戏读完,如果某个价值 (爱/恨、希望/绝望、自由/困境) 没有从正翻到负或从负翻到正,这场戏就没有存在理由 — 删或重写
  • 应用方式: 每写完一场戏,问 Mamet 三问: 角色想要什么? 障碍是什么? 价值转了吗? 不能三答 = 删
  • 局限: 纯氛围/节奏场 (如《刺客聂隐娘》) 不以价值翻转为核心; 短剧/微短剧的密度节奏不允许每场展开完整翻转
  • evidence: [T01-S001, T01-S014, T04-S007, T04-S028, T06-S002]

1.3 结构是工具不是紧身衣 (Structure as Diagnostic, Not Straitjacket)

(figures: Truby / McKee / Yorke vs Field / Snyder)

  • 一句话: 节拍表/三幕结构/英雄之旅是诊断已完成作品问题的工具,不是从空白页开始填空的模板 — 用模板填出来的故事可预测、同质化,用原理长出来的故事才有机
  • 应用方式: 入门用节拍表当脚手架快速建立结构感; 进阶用 Truby/McKee 视角交叉检验「这个节拍是从角色道德有机长出的还是硬塞的」;资深把模板扔掉,只留诊断能力
  • 局限: 纯工业化流水线作品 (如中国平台定制甜宠/古偶量产) 实际上就是按模板填空的,这个心智模型对量产场景不适用 — 但量产本身是工艺退化,不是方法论选择
  • evidence: [T01-S006, T01-S010, T04-S005, T04-S009, T04-S013]

1.4 对白即动作与潜台词 (Dialogue as Action — Subtext Over Surface)

(figures: Mamet / Sorkin / 兰晓龙 / 刘和平 / McKee)

  • 一句话: 人物开口说的每一句话都是在做一件事 — 求爱、威胁、试探、遮掩 — 台词的意思永远不等于字面意思; 潜台词是对白的灵魂,把意思说透了的台词叫「太直白」(on the nose)
  • 应用方式: 写对白前先定义「这个人物此刻想从对方手里得到什么」,然后让他不直说。检测方法: 如果把一段对白的字面意思替换成另一种说法,场景的意义变了吗? 没变 = 有潜台词; 变了 = on the nose
  • 局限: 部分中国类型剧 (主旋律/说教性作品) 对白承担信息传达功能,潜台词空间被挤压; Sorkin 式高密度智性对白风格不适用于所有 register (安静、内敛的剧不需要)
  • evidence: [T01-S013, T01-S014, T04-S028, T04-S024, T06-S005]

1.5 类型是对观众的承诺 (Genre as Emotional Contract)

(figures: Snyder / 芦苇 / Truby / McKee / 邹静之)

  • 一句话: 类型不是题材标签 (不是「有枪就是动作片」),是对观众的情感承诺 — 选了惊悚类型,观众买票来看紧张/恐惧,你必须兑现; 颠覆类型前必须先掌握类型
  • 应用方式: 确定故事类型后,列出这个类型的必备场景 (obligatory scenes) 和观众期待,逐一检查是否兑现;缺一个 = 观众觉得「被骗了」。Snyder 的 10 类故事按情感而非题材分类是入门最好框架
  • 局限: 作者电影/文艺片/实验叙事不遵循类型合约逻辑 (如《穆赫兰道》); 中国「主旋律」类型的观众期待与市场类型片不同 (政治正确 > 情感承诺)
  • evidence: [T01-S010, T01-S022, T04-S013, T04-S007, T06-S018]

1.6 写作就是改写 (Writing Is Rewriting — 初稿是原材料)

(figures: Goldman / August / Mazin / 芦苇 / King / 编剧通识)

  • 一句话: 初稿不是作品,是原材料 — 所有好剧本都是改出来的,不是一气呵成的; 片方/平台的 notes 常是错的,但它指向的问题是真的 (note behind the note)
  • 应用方式: 写完初稿后按三层诊断: 结构层 (三幕/节拍位置) → 人物层 (弧光/want-need) → 场景层 (每场的价值转) → 对白层 (潜台词/声口)。越大层面的问题越先修。找到 note behind the note 而非照改
  • 局限: 微短剧/网大模式下初稿=终稿是常态 (时间和预算不允许); 改稿无限循环 = development hell,需要知道何时停
  • evidence: [T01-S019, T01-S022, T04-S025, T04-S031, T03-S002]

1.7 主题是被戏剧化的论证 (Theme as Dramatized Argument, Not Message)

(figures: Truby / McKee / Mazin / Egri / 刘和平)

  • 一句话: 好故事的主题不是「告诉观众什么道理」,而是通过让持对立价值观的人物碰撞来让观众自己得出结论 — 主题先行的作品变说教,主题隐入故事的作品才有力量
  • 应用方式: 写故事前问「这到底在讲什么」,得到一句掌控性理念/前提 (如「贪婪导致毁灭,因为...」),然后让主角和反派分别代表这个论证的正反面,用他们的冲突来呈现,不用旁白
  • 局限: 主旋律/命题作品的主题是预设的 (如「抗疫英雄」),创作空间在「怎么让预设主题好看」而非「要不要有主题先行」; 类型片有时不需要深层主题,情感体验本身就够
  • evidence: [T01-S006, T01-S019, T04-S009, T04-S003, T06-S009]

标准 Playbook

  1. 如果主角的想要和需要不冲突: 则人物立不住 — 回头给主角一个 flaw (致命缺陷) 和一个 wound (创伤),让外在目标和内在需要拉扯。案例: Walter White 想要 (给家人留钱) vs 需要 (被世界承认) evidence: [T04-S009, T01-S006]
  2. 如果一场戏读完没有价值翻转: 则删或重写 — 问 Mamet 三问 (角色想要什么/障碍/价值转了吗),三答不出就剪。案例: 《社交网络》每场存证会戏都翻转权力关系 evidence: [T04-S007, T01-S014]
  3. 如果收到改稿意见 (notes): 则不要照改 — 找 note behind the note (真问题而非表面指令)。片方说「第二幕太慢」,真问题可能是「主角目标不清」。案例: Craig Mazin 谈 Chernobyl notes evidence: [T01-S019, T04-S021]
  4. 如果 logline 用一句话说不清主角+冲突+风险: 则概念不够锋利 — 回头重想,不要急着写大纲。Logline 是压力测试。案例: Alien 的 logline 「太空中没人听到你的尖叫」 evidence: [T04-S021, T03-S004]
  5. 如果类型片缺必备场景: 则观众会觉得被骗 — 列出该类型的 obligatory scenes,逐一检查。悬疑必有「真相揭露」; 爱情必有「黑暗时刻+重逢」。案例: Snyder 《救猫咪》对 10 类的必备场景清单 evidence: [T04-S013, T01-S010]
  6. 如果对白太直白 (on the nose): 则加潜台词层 — 让角色说的和想的不一样。检测: 能不能给这段对白加括号写「(其实他想说的是...)」? 写不出 = 没潜台词。案例: 《教父》开场求婚礼恩情 evidence: [T04-S028, T06-S005]
  7. 如果故事结构出问题,优先查第一幕: 则激励事件位置或主角目标可能没定清 — 第二幕坍塌 (约 90% 编剧 自称难点, 业内估) 的根因多在第一幕。案例: McKee 讲「第二幕问题向第一幕找答案」 evidence: [T04-S007, T01-S001]
  8. 如果入行编剧: 则写 spec (原创) 而非等委托 — 无名编剧靠 spec 证明能力; 投 Black List/竞赛/agent,spec 是入场券。案例: Diablo Cody《朱诺》spec → Oscar evidence: [T03-S023, T02-S001]
  9. 如果做中国长剧集项目: 则审查判断前置到选题阶段 — 不要写完 40 集再发现拍不了;备案/立项/龙标是硬约束,提前评估题材红线。案例: 广电总局题材引导目录 evidence: [T03-S012, T06-S033]
  10. 如果想颠覆类型: 则先完全掌握该类型的惯例再打破 — 不懂规则的「颠覆」不是颠覆,是不专业。案例: 《惊声尖叫》(Scream) 先遵守再解构恐怖片 evidence: [T04-S013, T01-S022]

工具栈与选型决策树

3.1 必备工具 (7 件, 约 80% 编剧都用 — 业内估)

  1. Final Draft — 好莱坞行业标准剧本格式软件 40+ 年, .fdx 格式是产业通用语, 制片/经纪人/studios 默认收 evidence: [T02-S001]
  2. WriterDuet — 协作云端剧本软件, 免费层 + pro, 快速增长替代 Final Draft evidence: [T02-S002]
  3. Scrivener — 长稿组织/大纲 binder 模式, 不是剧本格式软件但所有编剧用于前期组织 evidence: [T02-S005]
  4. Index Cards (物理或数字) — 传统 beat/scene 排列工具, 从 Syd Field 开始至今, 不可替代的具象思维工具 evidence: [T02-S008]
  5. The Black List — spec 剧本评估平台 + 年度榜单, 入行核心通道 evidence: [T02-S011]
  6. Fade In Pro — 轻量替代 Final Draft ($79 买断, 非订阅), 独立编剧首选 evidence: [T02-S003]
  7. Highland Pro — John August 公司 Mac 原生, Fountain 纯文本格式, 无锁定风险 evidence: [T02-S004]

3.2 场景特化 (7 类, 按子方向不同)

  • 长篇电影剧作: Final Draft/Fade In + Scrivener (大纲) + Index Cards
  • 美剧编剧室: WriterDuet (协作) + Final Draft (最终交付)
  • 中国长剧集: Word/WPS (中文格式无标准化软件) + 平台投稿系统 (爱优腾芒)
  • 微短剧: 轻量工具 (Google Docs/Notion) + 平台创作后台 (快手/抖音)
  • IP改编: Scrivener (source material 组织) + 节拍表/大纲对比工具
  • 竞赛投递: Coverfly(已关闭→Stage 32 接力)/Black List/InkTip + 格式自检
  • AI 辅助: ChatGPT/Claude (brainstorm, WGA: AI 不可作署名作者) / Sudowrite (prose-focused)

3.3 新兴工具 (5 件, 近 12 月出现或起势)

  1. Arc Studio Pro — 云端新兴剧本软件 (免费层), 增长快 evidence: [T02-S006]
  2. StudioBinder (剧本功能) — 2025 新增免费剧本写作, 从制作管理扩展 evidence: [T02-S010]
  3. AI Coverage 服务 — 用 AI 做首轮 coverage, 成本低但质量争议 evidence: [T02-S039]
  4. Sudowrite — AI 辅助写作, Muse 模型, $10-44/月 evidence: [T02-S038]
  5. StoryPeer — Coverfly 关闭后社区型替代, 2025-2026 新兴

3.4 避坑清单 (5 条)

  1. ❌ 不要格式不规范就投: 非 .fdx 或 pro-level 格式, coverage reader 第一页就 pass
  2. ❌ 不要混淆 Scrivener 和 Final Draft: Scrivener 做大纲, Final Draft 做格式化稿, 不是替代关系
  3. ❌ 不要让 AI 写完整 draft 署你名: WGA 2023 协议明确 AI 不可作署名作者; AI 辅助 brainstorm/research 可以, 替代创作不行
  4. ❌ 不要依赖单一竞赛平台: Coverfly 2025 关闭的教训, 多平台投
  5. ❌ 不要用中文 Word 模板写好莱坞格式 spec: 格式是行业语言, 不规范 = 不专业

工作流 / Pipeline

4.1 工作流清单 (10 个, 按 decay 分组)

Low decay (工艺稳定):

1. 从零写一个电影剧本

入门 SOP: 1. 确定前提/Logline → 2. 写 Treatment 故事大纲 → 3. 做 Beat Sheet 节拍表 → 4. 写 Step Outline 分场大纲 → 5. 写初稿 (vomit draft) → 6. 自我诊断+改稿

  • 资深路径: 跳过 Step 3 beat sheet / 优化 Step 5 边写边改 / 额外 Step 0 拉片+采风 evidence: [T03-S001, T03-S003]

4. 改稿与开发

入门 SOP: 1. 读 notes 两遍不急回应 → 2. 诊断 (结构→人物→场景分层) → 3. 找 note behind the note → 4. Draft revision plan → 5. Rewrite → 6. Polish → 7. Table read

  • 资深路径: 跳过 Step 2 机械诊断 / 优化 Step 3 秒定真问题 / 额外: 预判下一轮 notes evidence: [T03-S002, T03-S014]

5. IP 改编

入门 SOP: 1. 精读原著找戏剧脊柱 → 2. 列人物关系+核心冲突 → 3. 压缩/扩展+建场景结构 → 4. 写改编 treatment → 5. 初稿 → 6. 与原作对标复查

  • 资深路径: 跳过 Step 6 逐章对标 / 优化 Step 1 直觉找到脊柱不需逐章 / 额外: 提前与 IP 方确认改编自由度 evidence: [T03-S006, T03-S010]

9. 类型片惯例与必备场景

入门 SOP: 1. 确定故事类型 (Snyder 10 类) → 2. 列该类型必备场景 → 3. 写节拍时逐一嵌入 → 4. 初稿后用必备场景清单诊断 → 5. 补缺失的必备场景

  • 资深路径: 跳过 Step 2 显式列表 / 优化: 同时用多类型交叉验证 / 额外: 有意识颠覆 1-2 个惯例 evidence: [T03-S020, T03-S007]

10. 人物发展与对白工艺

入门 SOP: 1. 写人物小传 (三维: 生理/社会/心理) → 2. 定 want/need/flaw → 3. 设弧光轨迹 → 4. 写对白 (先写直白版再加潜台词层) → 5. 盲测: 遮住角色名看能否分辨谁在说

  • 资深路径: 跳过 Step 1 完整小传 / 优化 Step 4 直接写潜台词 / 额外: 为每个角色建独立词频表 evidence: [T03-S010, T03-S006]

Medium decay:

2. 美剧编剧室运作

入门 SOP: 1. Showrunner 定义 series bible (角色/世界/语气) → 2. 编剧室 break the season (全季弧线/集弧线上墙) → 3. 分配 episode (每人写 1-2 集) → 4. 个人 outline → 全室 rewrite → 5. Table read → 6. Production draft

  • 资深路径: 跳过 Step 3 等分配 (资深人主动抢 bottle episode 或 finale) / 优化 Step 4 outline 直接带对白不只结构 / 额外: 预判制片/演员 notes 在 draft 阶段前置化解 evidence: [T03-S008, T03-S020]

3. 中国长剧集编剧流程

入门 SOP: 1. 故事大纲 (3-5 页) → 2. 人物小传 (主要 5-8 个角色) → 3. 分集大纲 (30-40 集各 500-800 字, 业内估) → 4. 剧本 (逐集) → 5. 改稿 (处理意见/平台 notes) → 6. 审查/备案/立项

  • 资深路径: 跳过 Step 2 独立小传 (人物在大纲阶段同步立) / 优化 Step 5 主动控制注水压力 (拒绝为集数拉长牺牲叙事密度) / 额外: Step 0 前置审查评估 (题材红线判断比写完再评省半年) evidence: [T03-S012, T03-S013]

6. Spec 投递与评估

入门 SOP: 1. 完成 spec 剧本 + polish → 2. 写 query letter / logline → 3. 投 Black List ($30 评估) + 竞赛 (Nicholl/Austin/PAGE) → 4. 收 coverage → 5. 根据反馈改稿 → 6. 被 consider/recommend → 联系 agent/manager

  • 资深路径: 跳过 Step 3 广撒网竞赛 (靠人脉/推荐直接进 manager inbox) / 优化 Step 4 筛选哪些 coverage 值得听 / 额外: 同时维护 2-3 个 active spec 而非只投一个 evidence: [T03-S023, T02-S011]

8. 中国审查与备案流程

入门 SOP: 1. 题材自审 (广电总局题材引导目录) → 2. 故事大纲报备 → 3. 完成剧本 → 4. 提交审查 → 5. 修改意见 → 6. 获龙标/平台上线许可

  • 资深路径: 跳过 Step 1 查目录 (内化红线) / 优化 Step 2 大纲阶段就与审查方沟通方向 / 额外: 预设「B 方案」(审查打回时的替代结局/情节) evidence: [T03-S012, T06-S033]

High decay:

7. 微短剧/竖屏短剧创作

入门 SOP: 1. 爽点设计 (核心情绪 hook: 逆袭/复仇/甜宠) → 2. 80-100 集结构 (每集 90 秒, 1-3 分钟) → 3. 钩子+反转密度 (每集末尾必有 cliffhanger) → 4. 充值点设计 (第 10-15 集首个付费墙) → 5. 批量产出 → 6. 数据反馈迭代

  • 资深路径: 跳过 Step 4 机械充值点 (用叙事自然断点而非硬卡) / 优化 Step 1 在爽点套路中加人物弧光 (让微短剧有工艺升级) / 额外: 同时运营 3-5 个不同类型项目分散风险 evidence: [T03-S025, T06-S040]

表达 DNA

5.1 高频用语 / 黑话 (top 15)

logline / beat / act break / cold open / inciting incident / midpoint / climax / controlling idea / subtext / on the nose / show don't tell / enter late leave early / the gap / character arc / vomit draft / 人物弧光 / 潜台词 / 分集大纲 / 过审 / 注水 / 飞页

5.2 严肃讨论 vs 轻松吐槽 Register

  • 严肃: 「这场戏的 value charge 没有转, controlling idea 不清晰, B story 跟 A story 的 thematic tie 断了」
  • 轻松: 「第二幕写死了吧? 节拍全乱了, 这 midpoint 是假的吧」「这台词太直白了 on the nose 得要死」
  • 中国: 「这戏水了 (注水)」「人物立不立得住」「这场没戏, 没有信息量」「飞页了 (未通知改稿)」

5.3 外行破绽 (outsider-tell top 10)

  1. 把「story」和「plot」混用 (McKee 严格区分: plot 是事件序列, story 含 values/meaning)
  2. 说「剧情」指的是「故事」(中文混用; 行内人区分剧情=情节线 vs 故事=完整结构+主题+人物)
  3. 用「主角」指所有角色 (行内: protagonist 只有一个, 其他是 supporting/ensemble)
  4. 认为「好对白 = 聪明的台词」(行内: 好对白 = 有潜台词, 不一定聪明)
  5. 把 screenplay 和 script 当同义词 (screenplay 特指电影剧本, script 更泛)
  6. 说「这个故事没有结构」(行内: 所有故事都有结构, 只是你没看出来)
  7. 混淆「改编」和「翻拍」(adaptation 是媒介转换, remake 是同媒介重做)
  8. 认为「过审」只是内容审查 (实际包括备案/立项/龙标/平台审核多层)
  9. 把 spec script 说成「投稿」(spec = 编剧主动写的原创, 不是「投稿」这么被动)
  10. 认为编剧室 = 编剧团 (writers' room 是美剧特有的 collaborative breaking 制度, 中国的「编剧团」多是分集承包制)

5.4 厂商话术拒绝

  • ❌ 「AI 写剧本只需要 5 分钟」→ 行内人拒绝: AI 可以辅助 brainstorm, 但编剧的核心工作是判断+选择, AI 替代不了
  • ❌ 「用我们的模板就能写出好莱坞大片」→ 行内人拒绝: 模板是工具不是魔法, 救猫咪填空 ≠ 好剧本
  • ❌ 「我们的 coverage 用 AI 评估更客观」→ 行内人拒绝: coverage 需要人类 reader 的品味判断, AI coverage 仍不被业内认可
  • ❌ 「剧本软件决定剧本质量」→ 行内人拒绝: Goldman 用打字机写 Butch Cassidy, 工具不决定工艺

5.5 对话样本库 (industry voice 实战语料)

5.5.1 客户版 (面客解释/教育)

  • 「结构不是往框里填东西。好比盖房子,你先得知道住什么人、几口人、干什么用,然后才画图。节拍表是你画完图之后的验收清单,不是设计图本身。」(source: T01-S006, 转述, 客户场景: 编剧教学)
  • 「类型不是拍什么题材,是给观众什么感受。你选了惊悚,观众买票来被吓,你没吓到他就是违约。类型是合同。」(source: T01-S010, 转述, 客户场景: 制片人沟通)

5.5.2 同业版 (私下/内训/同业访谈)

  • 「Every scene is a negotiation. Somebody wants something, and somebody else is in the way. If that's not happening, why am I watching?」(source: T01-S014, 原话, Mamet)
  • 「The note is always wrong. The problem the note is trying to fix is always real.」(source: T01-S019, 转述, Mazin on Scriptnotes)
  • 「这戏水了,第二幕全在晃,主角什么都不想要。」(source: T01-S022, 推断, 中国编剧内部讨论)

5.5.3 监管/专业版 (公开谈行业/工艺)

  • 「Intention and obstacle — every great scene is someone who desperately wants something and someone who's preventing them from getting it.」(source: T01-S013, 原话, Sorkin MasterClass)
  • 「中国电影需要类型自觉。我们不缺故事,缺的是把故事放进类型框架里的工艺训练。」(source: T01-S022, 转述, 芦苇)
  • 「The audience doesn't know what they want. Your job is not to give them what they want — it's to give them what they didn't know they needed.」(source: T01-S030, 转述, Rhimes)

5.5.4 反例版 (资深人绝不会这样说)

  • 「主角不用有缺点观众才喜欢」→ why 反例: 无 flaw 的角色是 Mary Sue, 没弧光可走 (source: T06 outsider-tell)
  • 「台词把意思说清楚就行潜台词太装」→ why 反例: 没潜台词 = on the nose = 外行第一特征 (source: T06-S005)
  • 「抄个爆款换皮最快反正改改桥段」→ why 反例: 系统性剽窃有法律判例 (琼瑶诉于正 2014 判赔, source: T01-R1)
  • 「节拍表填空照写就是好剧本」→ why 反例: Truby/McKee 都明确反对, 滥用 beat sheet 导致同质化 (source: T04-S009)

voice_confidence: medium (西方 figures 原话丰富 = high; 中国 figures 多转述/推断 = low; 综合 medium)

质量基准 + 反模式

6.1 质量基准 (什么算好)

  1. 人物弧光完整: want/need 冲突清晰, 弧光在高潮处通过选择完成 (正弧/负弧/平弧都行, 但必须有意识设计)
  2. 每场戏有价值转: 场景不能只传递信息, 必须转变某个价值/关系/权力
  3. 对白有潜台词: 盲测遮住角色名能分辨谁在说; 关键场景的台词字面意思 ≠ 实际意思
  4. 结构有呼吸: 节奏有张有弛, 不是一条线到底; 第二幕中点有明确的方向转变
  5. logline 一句话能说清: 能在 25 字 (中文) 或 1 句话 (英文) 说清主角+冲突+风险

6.2 反模式 (外行/入门常犯错, 10 条)

  1. 跳过大纲直接写初稿 → 第二幕必塌 (约 90% 入门编剧的经历, 业内估) evidence: [T03-S001]
  2. beat sheet 填空当创作 → 机械公式化, Truby/McKee 反复批判 evidence: [T04-S009]
  3. 对白太直白 (on the nose) → 没潜台词 = 外行第一特征 evidence: [T06-S005]
  4. 主角是 Mary Sue (无缺陷) → 没弧光, 没内在张力 evidence: [T04-S009]
  5. 不改稿 / 一稿定稿 → 编剧通识: 所有好剧本都是改出来的 evidence: [T01-S019]
  6. 照改 notes 而非找真问题 → 改出一版更差的稿子 evidence: [T03-S014]
  7. 格式不规范就投 → coverage reader 第一页 pass evidence: [T02-S001]
  8. 中国项目不前置审查判断 → 写完 40 集发现拍不了 evidence: [T03-S012]
  9. 混淆类型惯例和「万能模板」 → 类型是诊断工具不是生产线 evidence: [T04-S013]
  10. 抄袭他人桥段/系统性剽窃 → 有明确法律判例 (于正 2014 / 郭敬明 2006), 不是「灰色地带」evidence: [T01-R1]

智识谱系

7.1 主要流派

  1. 古典情节派 (Aristotle → 情节优先, 突转/发现/净化是戏剧本质)
    • 当前代表: 学术传统, McKee 部分继承
  2. 现代结构教学派 (Syd Field → 三幕范式/页码坐标 → Blake Snyder 精细化 → 15 拍节拍表)
    • 当前代表: Save the Cat! 社区 (Jessica Brody), 大量编剧课程
  3. 人物有机结构派 (Egri → 人物决定结构 → Truby 22 步/道德论证)
    • 当前代表: John Truby 及其写作课
  4. 价值/哲学派 (McKee → 掌控性理念/鸿沟/价值转变, 反公式但也非纯人物)
    • 当前代表: Robert McKee 及 Story Seminar
  5. 神话/原型派 (Joseph Campbell → Christopher Vogler 英雄之旅 → Dan Harmon 故事环)
    • 当前代表: Dan Harmon (TV 实用化)
  6. 对白/场景大师派 (David Mamet 极简策略性 vs Aaron Sorkin 高密度智性)
    • 两极: Mamet (every scene is a negotiation) vs Sorkin (intention and obstacle)
  7. 院校教学派 (Frank Daniel 八序列法 → USC/Columbia 传统)
    • 当前代表: USC SCA MFA, 大学编剧课程
  8. 行业智慧/实务派 (Goldman「Nobody knows anything」→ August/Mazin Scriptnotes → Franklin Leonard Black List)
    • 当前代表: John August, Craig Mazin, Franklin Leonard
  9. 中国编剧工艺自觉派 (芦苇 类型自觉 + 刘和平 历史正剧巅峰 + 邹静之 编剧维权)
    • 当前代表: 芦苇, 刘和平, 宋方金 (维权/批评)
    • 核心张力: 西方正典工艺训练 vs 导演中心制下编剧地位弱 + 审查约束 + 流量资本逻辑

7.2 核心分歧 (保留不抹平)

  1. 公式派 vs 反公式派: Field/Snyder (结构=外加模板) vs Truby/McKee (结构=有机生长) — 编剧理论最大内部张力
  2. 情节优先 vs 人物优先: Aristotle vs Egri — 2400 年未解,实务中好作品两端都有
  3. 英雄之旅: 普适 vs 被滥用: Vogler/Harmon (一个原型适用多数) vs Truby/Yorke (非所有故事是旅程)
  4. 美剧 vs 中国编剧权力结构: 美国 showrunner 权力最大 vs 中国导演中心制, 编剧地位结构性不同

诚实边界

  1. 信息截止 2026-05-23: 工具/工作流模块衰减最快 (特别是微短剧/AI 辅助/竞赛平台生态), 建议每 6 个月 update
  2. Canon 西方压倒性: 编剧 craft 正典约 90%+ 来自西方英文圈 (Aristotle→McKee→Truby→Snyder 谱系); 中文原生工艺正典仅苏牧/芦苇/张先 3 本 + 大量西方译本。这是真实结构, 不是数据缺陷
  3. 中国 voice DNA 结构性薄: 中国编剧公开 voice 大量在 B站/公众号/知乎/微博 (blacklist/reference 平台), 导致 zh-CN 维度一手逐字原话极少; 中国 figure 的 voice_samples 多为转述/推断。如需中国编剧视角深度, 建议直接看芦苇《电影编剧的秘密》+ 刘和平署名长访谈
  4. 工具栈稳态 + AI 快变: 核心工具 (Final Draft/Scrivener) 40+ 年稳定; 但 AI 辅助编剧工具和 WGA 的 AI 政策正在 2025-2026 快速变化 (WGA 2026 谈判将重塑 AI 边界)
  5. 竞赛/平台生态动荡: Coverfly/ScreenCraft/WeScreenplay 2025 关闭, spec 投递生态正重组
  6. 公开信息偏差: 成功编剧大量发声 (McKee/Sorkin/Mazin 都有 MasterClass/podcast), 但商业失败/审查被毙/维权败诉的案例公开极少; 幸存者偏差严重
  7. 微短剧工艺争议: 微短剧的「爽点工业化」是否是编剧工艺正在被讨论, 本 skill 收录其 SOP 但不为其工艺质量背书

Time-decay Registry

This skill's modules decay at different speeds. Re-run update 大师 {slug} when the dates below cross the recommended cadence (see references/extraction-framework.md § 八).

Modulelast_updateddecay_riskRecommended refresh cadence
Mental modelslast_updated: 2026-05-23decay_risk: low1-2 years
Standard playbooklast_updated: 2026-05-23decay_risk: low6-12 months
Tool stacklast_updated: 2026-05-23decay_risk: high3-6 months
Workflows / pipelinelast_updated: 2026-05-23decay_risk: high3-6 months
Expression DNAlast_updated: 2026-05-23decay_risk: low6-12 months
Sources (Track 5)last_updated: 2026-05-23decay_risk: medium6 months
Glossary / standards / regulationslast_updated: 2026-05-23decay_risk: medium6 months (regulations may force sooner)
Intellectual genealogylast_updated: 2026-05-23decay_risk: low1-2 years
Honest boundarieslast_updated: 2026-05-23decay_risk: lowre-assess each refresh

last_updated values reflect the synthesis date. Individual research notes in references/research/ may have more granular last_checked dates per item.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

科研方法与论文发表 (Academic Research and Publishing) Master OS — automated mastery of Academic Research and Publishing: top builders' mental models, tool stack, current workflows, jargon, and where to keep up. Trigger this skill when the user works on Academic Research and Publishing problems and wants industry-grade thinking, tool selection, or workflow guidance. 触发词:「科研方法」「论文发表」「学术写作」「论文写作」「科研写作」

日本語の概要は準備中です。原文の説明を表示しています。

swaylq/master-skill1492026年9月6日 更新

广告外包公司绩效管理 (广告外包公司绩效管理 (Ad Agency / Marketing-Outsourcing Firm Performance Management) — 广告/营销外包公司(乙方 agency:创意热店/4A/本土广告公司/传媒公司/数字营销代理/MCN/设计工作室/内容外包)的『怎么做绩效管理』职业认知操作系统,面向老板/合伙人、运营总监(COO)、HR/HRBP、项目总监、team lead、以及想搭绩效体系的从业者。覆盖: (a) 第一性张力 — 绩效管理的核心世界观分歧: **创意/知识工作『不可量化』 vs 『必须可考核』(创意难归因、难标准化,硬套 KPI 杀创意;但不考核养懒人——这是本行绩效的元矛盾) ⇄ 结果导向(KPI/业绩/回款) vs 过程导向(OKR/能力/成长) ⇄ 个人绩效 vs 团队协作(广告是集体产物,创意-AE-媒介-设计协作,难拆到个人) ⇄ 强激励(提成制:多劳多得但抢单/短视/不协作/吃老客户) vs 稳定(固定+奖金:公平但大锅饭) ⇄ 规模/流水 vs 利润/人效(低价接单冲规模 vs 保毛利)**; (b) 绩效体系设计 — KPI(关键指标法)/OKR(目标与关键成果)/BSC(平衡计分卡)/360 环评/MBO(目标管理)/KSF(关键成功因子)/积分制/计件提成;考核周期(月/季/年+项目制);绩效面谈、绩效改进 PIP、强制分布(271/活力曲线)争议、绩效结果应用(调薪/晋升/淘汰/奖金); (c) 分岗位绩效(本行难点,各岗量化差异极大) — 创意(文案/美术/创意总监:作品质量、获奖、比稿命中率——最难量化)、AE/客户执行(客户满意度、续约、回款、增购、客诉)、媒介/投放(可量化:ROI/ROAS/CPM/转化——最易量化)、策略/策划(方案通过率、洞察)、设计(产能/返稿率/品质)、项目管理(交付准时率、利润率)、销售/BD(新签、回款、毛利); (d) 项目与客户维度绩效(agency 特有,专业服务公司逻辑) — 项目利润率(收入-人力成本-外部成本)、工时制与人效(billability 可计费工时率/utilization 利用率/realization 实现率,借鉴咨询律所)、人均产值/人均毛利、客户续约率与 LTV、客户集中度风险、比稿投入产出、改稿次数失控; (e) 激励与分配机制 — 提成制(按签单/毛利/回款抽成)、项目奖金分成(如何在创意-AE-媒介间分)、年终奖包、合伙人制/事业合伙人/工作室 studio 制/阿米巴、股权期权、创意人才的非物质激励(署名/获奖/成长); (f) 人效、成本与文化 — 人效比(人力成本占收入比,业内估计通常 40-55%,非普适)、毛利率、加班/血汗工厂文化批判(乙方 996/改稿到凌晨)、人才流失率与保留、招聘与培养、组织架构(客户组 vs 职能组 vs 项目制矩阵); (g) 外包/乙方特有张力对绩效的侵蚀 — 甲乙方权力不对等(甲方压价、无限改稿、比稿白嫖创意、账期长垫款重回款难)、低价内卷与红海、项目制不确定性(旺淡季、大客户依赖)、创意工时不可控如何计入绩效、外包 vs 自建团队的成本对比; (h) 争议/批判 — KPI 量化派 vs 创意不可考核派、提成制养蛊 vs 团队协作、强制分布/末位淘汰是否适合小 agency、工时制(billable hours)被咨询律所验证但绑架员工健康、『向甲方看齐做 OKR』水土不服、绩效主义扼杀创意(索尼绩效主义之殇式批判)、乙方绩效本质是『用考核对冲甲方不确定性』的转嫁; (i) 流派/思想谱系 — 专业服务公司管理派(David Maister《Managing the Professional Service Firm》PSF 圣经:leverage 杠杆/utilization 利用率/billing;借鉴咨询律所会计所) vs agency 专家经营派(David C. Baker《The Business of Expertise》+2Bobs podcast、Blair Enns《Win Without Pitching》《Pricing Creativity》定位与定价) vs OKR/目标管理派(John Doerr《Measure What Matters》、Google/Intel OKR、字节/飞书 OKR 本土化) vs 中式绩效工具派(KSF 薪酬全绩效/积分制/阿米巴/华为绩效体系被广告公司借用) vs 本土 4A/广告公司实战派(省广/蓝色光标/华扬联众/利欧 等上市广告公司管理 + 创意热店工作室制)。不含: 广告投放/信息流/SEM 的技术操作本身、甲方市场部/品牌方内部绩效、通用 HR 招聘薪酬全流程(仅取与绩效相关部分)、纯财务会计做账、MCN 达人孵化运营细节。) Master OS — automated mastery of 广告外包公司绩效管理 (Ad Agency / Marketing-Outsourcing Firm Performance Management) — 广告/营销外包公司(乙方 agency:创意热店/4A/本土广告公司/传媒公司/数字营销代理/MCN/设计工作室/内容外包)的『怎么做绩效管理』职业认知操作系统,面向老板/合伙人、运营总监(COO)、HR/HRBP、项目总监、team lead、以及想搭绩效体系的从业者。覆盖: (a) 第一性张力 — 绩效管理的核心世界观分歧: **创意/知识工作『不可量化』 vs 『必须可考核』(创意难归因、难标准化,硬套 KPI 杀创意;但不考核养懒人——这是本行绩效的元矛盾) ⇄ 结果导向(KPI/业绩/回款) vs 过程导向(OKR/能力/成长) ⇄ 个人绩效 vs 团队协作(广告是集体产物,创意-AE-媒介-设计协作,难拆到个人) ⇄ 强激励(提成制:多劳多得但抢单/短视/不协作/吃老客户) vs 稳定(固定+奖金:公平但大锅饭) ⇄ 规模/流水 vs 利润/人效(低价接单冲规模 vs 保毛利)**; (b) 绩效体系设计 — KPI(关键指标法)/OKR(目标与关键成果)/BSC(平衡计分卡)/360 环评/MBO(目标管理)/KSF(关键成功因子)/积分制/计件提成;考核周期(月/季/年+项目制);绩效面谈、绩效改进 PIP、强制分布(271/活力曲线)争议、绩效结果应用(调薪/晋升/淘汰/奖金); (c) 分岗位绩效(本行难点,各岗量化差异极大) — 创意(文案/美术/创意总监:作品质量、获奖、比稿命中率——最难量化)、AE/客户执行(客户满意度、续约、回款、增购、客诉)、媒介/投放(可量化:ROI/ROAS/CPM/转化——最易量化)、策略/策划(方案通过率、洞察)、设计(产能/返稿率/品质)、项目管理(交付准时率、利润率)、销售/BD(新签、回款、毛利); (d) 项目与客户维度绩效(agency 特有,专业服务公司逻辑) — 项目利润率(收入-人力成本-外部成本)、工时制与人效(billability 可计费工时率/utilization 利用率/realization 实现率,借鉴咨询律所)、人均产值/人均毛利、客户续约率与 LTV、客户集中度风险、比稿投入产出、改稿次数失控; (e) 激励与分配机制 — 提成制(按签单/毛利/回款抽成)、项目奖金分成(如何在创意-AE-媒介间分)、年终奖包、合伙人制/事业合伙人/工作室 studio 制/阿米巴、股权期权、创意人才的非物质激励(署名/获奖/成长); (f) 人效、成本与文化 — 人效比(人力成本占收入比,业内估计通常 40-55%,非普适)、毛利率、加班/血汗工厂文化批判(乙方 996/改稿到凌晨)、人才流失率与保留、招聘与培养、组织架构(客户组 vs 职能组 vs 项目制矩阵); (g) 外包/乙方特有张力对绩效的侵蚀 — 甲乙方权力不对等(甲方压价、无限改稿、比稿白嫖创意、账期长垫款重回款难)、低价内卷与红海、项目制不确定性(旺淡季、大客户依赖)、创意工时不可控如何计入绩效、外包 vs 自建团队的成本对比; (h) 争议/批判 — KPI 量化派 vs 创意不可考核派、提成制养蛊 vs 团队协作、强制分布/末位淘汰是否适合小 agency、工时制(billable hours)被咨询律所验证但绑架员工健康、『向甲方看齐做 OKR』水土不服、绩效主义扼杀创意(索尼绩效主义之殇式批判)、乙方绩效本质是『用考核对冲甲方不确定性』的转嫁; (i) 流派/思想谱系 — 专业服务公司管理派(David Maister《Managing the Professional Service Firm》PSF 圣经:leverage 杠杆/utilization 利用率/billing;借鉴咨询律所会计所) vs agency 专家经营派(David C. Baker《The Business of Expertise》+2Bobs podcast、Blair Enns《Win Without Pitching》《Pricing Creativity》定位与定价) vs OKR/目标管理派(John Doerr《Measure What Matters》、Google/Intel OKR、字节/飞书 OKR 本土化) vs 中式绩效工具派(KSF 薪酬全绩效/积分制/阿米巴/华为绩效体系被广告公司借用) vs 本土 4A/广告公司实战派(省广/蓝色光标/华扬联众/利欧 等上市广告公司管理 +

日本語の概要は準備中です。原文の説明を表示しています。

swaylq/master-skill1492026年9月6日 更新

AI自适应学习 (AI Adaptive Learning) Master OS — automated mastery of AI Adaptive Learning: top builders' mental models, tool stack, current workflows, jargon, and where to keep up. Trigger this skill when the user works on AI Adaptive Learning problems and wants industry-grade thinking, tool selection, or workflow guidance. 触发词:「自适应学习」「智能诊断」「自适应题库」「学习路径规划」「知识追踪」

日本語の概要は準備中です。原文の説明を表示しています。

swaylq/master-skill1492026年9月6日 更新

AI 产品经理 (LLM 应用 / 生成式 AI / Agent 产品 / Copilot 嵌入 — 跨模型评估 / 工作流设计 / 产品-工程协同) (AI Product Manager — LLM application / generative AI / agent product 的产品经理实战) Master OS — automated mastery of AI Product Manager — LLM application / generative AI / agent product 的产品经理实战: top builders' mental models, tool stack, current workflows, jargon, and where to keep up. Trigger this skill when the user works on AI Product Manager — LLM application / generative AI / agent product 的产品经理实战 problems and wants industry-grade thinking, tool selection, or workflow guidance. 触发词:「AI 产品经理」「AI PM」「LLM 产品」「Generative AI product」「RAG 产品」

日本語の概要は準備中です。原文の説明を表示しています。

swaylq/master-skill1492026年9月6日 更新

AI 短剧 (AI 微短剧) (AI 短剧 (AI 微短剧) (从业者视角) — 用 AIGC 工具(AI 视频生成 / AI 脚本 / AI 生图 / AI 配音 / 对口型 / 角色一致性)制作竖屏微短剧并变现的新兴行业。覆盖: (a) 创作工作流 — 选题→剧本/分镜→AI 生图→图生视频(首尾帧/运镜/参考)→配音/对口型→剪辑/调色→成片,其中角色/场景一致性是第一性难题; (b) 工具栈 — 视频生成(可灵 Kling / 即梦 Seedance / 海螺 Hailuo / Vidu / Runway Gen / OpenAI Sora / Pika / Luma / Pixverse)、脚本(LLM)、生图(Midjourney/SD/即梦)、配音(TTS)、对口型、剪辑(剪映/CapCut); (c) 商业/变现 — 国内付费投流(IAP + ROI + 平台分账, 抖音/快手/红果/微信小程序短剧)与海外订阅(ReelShort/DramaBox/ShortMax/GoodShort)两套打法、品牌定制、成本结构; (d) 合规 — 中国《人工智能生成合成内容标识办法》(2025-09 强制 AI 标识)、广电微短剧备案与分类分层审核、版权与肖像权。主战场 = 抖音/快手/视频号/B站 + 海外短剧 App。学派分歧: 纯 AI vs AI+实拍混合、国内投流 vs 海外订阅、视频工具派之争、「AI 短剧是真需求还是伪风口」。不含: 传统实拍短剧制作、通用影视编剧(已有 scriptwriting skill)、短视频信息流投放(已有 short-video-ads skill)、纯 AI 视频底层技术研发。) Master OS — automated mastery of AI 短剧 (AI 微短剧) (从业者视角) — 用 AIGC 工具(AI 视频生成 / AI 脚本 / AI 生图 / AI 配音 / 对口型 / 角色一致性)制作竖屏微短剧并变现的新兴行业。覆盖: (a) 创作工作流 — 选题→剧本/分镜→AI 生图→图生视频(首尾帧/运镜/参考)→配音/对口型→剪辑/调色→成片,其中角色/场景一致性是第一性难题; (b) 工具栈 — 视频生成(可灵 Kling / 即梦 Seedance / 海螺 Hailuo / Vidu / Runway Gen / OpenAI Sora / Pika / Luma / Pixverse)、脚本(LLM)、生图(Midjourney/SD/即梦)、配音(TTS)、对口型、剪辑(剪映/CapCut); (c) 商业/变现 — 国内付费投流(IAP + ROI + 平台分账, 抖音/快手/红果/微信小程序短剧)与海外订阅(ReelShort/DramaBox/ShortMax/GoodShort)两套打法、品牌定制、成本结构; (d) 合规 — 中国《人工智能生成合成内容标识办法》(2025-09 强制 AI 标识)、广电微短剧备案与分类分层审核、版权与肖像权。主战场 = 抖音/快手/视频号/B站 + 海外短剧 App。学派分歧: 纯 AI vs AI+实拍混合、国内投流 vs 海外订阅、视频工具派之争、「AI 短剧是真需求还是伪风口」。不含: 传统实拍短剧制作、通用影视编剧(已有 scriptwriting skill)、短视频信息流投放(已有 short-video-ads skill)、纯 AI 视频底层技术研发。: top builders' mental models, tool stack, current workflows, jargon, and where to keep up. Trigger this skill when the user works on AI 短剧 (AI 微短剧) (从业者视角) — 用 AIGC 工具(AI 视频生成 / AI 脚本 / AI 生图 / AI 配音 / 对口型 / 角色一致性)制作竖屏微短剧并变现的新兴行业。覆盖: (a) 创作工作流 — 选题→剧本/分镜→AI 生图→图生视频(首尾帧/运镜/参考)→配音/对口型→剪辑/调色→成片,其中角色/场景一致性是第一性难题; (b) 工具栈 — 视频生成(可灵 Kling / 即梦 Seedance / 海螺 Hailuo / Vidu / Runway Gen / OpenAI Sora / Pika / Luma / Pixverse)、脚本(LLM)、生图(Midjourney/SD/即梦)、配音(TTS)、对口型、剪辑(剪映/CapCut); (c) 商业/变现 — 国内付费投流(IAP + ROI + 平台分账, 抖音/快手/红果/微信小程序短剧)与海外订阅(ReelShort/DramaBox/ShortMax/GoodShort)两套打法、品牌定制、成本结构; (d) 合规 — 中国《人工智能生成合成内容标识办法》(2025-09 强制 AI 标识)、广电微短剧备案与分类分层审核、版权与肖像权。主战场 = 抖音/快手/视频号/B站 + 海外短剧 App。学派分歧: 纯 AI vs AI+实拍混合、国内投流 vs 海外订阅、视频工具派之争、「AI 短剧是真需求还是伪风口」。不含: 传统实拍短剧制作、通用影视编剧(已有 scriptwriting skill)、短视频信息流投放(已有 short-video-ads skill)、纯 AI 视频底层技术研发。 problems and wants industry-grade thinking, tool selection, or workflow guidance. 触发词:「ai 短剧」「ai 微短剧」「ai short drama」「ai 短剧制作」「ai 短剧工作流」

日本語の概要は準備中です。原文の説明を表示しています。

swaylq/master-skill1492026年9月6日 更新

AIGC 创作工作流 (AIGC 创作工作流 / AIGC Creative Workflow — 以「生成式 AI 视觉创作(图像 + 视频)的工程化创作流水线」为对象的认知操作系统:从「需求/概念(要做什么画面/镜头/风格)→ 模型与工具选型(Midjourney / Stable Diffusion·SDXL·Flux / ComfyUI / 可灵 Kling / 即梦 Jimeng / Runway / Vidu / Sora)→ Prompt 与参数设计(提示词工程、风格描述、负向、种子、CFG/步数/采样器、参考图)→ 可控生成(ControlNet / LoRA / IPAdapter / inpainting / 图生图 / 局部重绘 / 区域控制)→ 批量出图与筛选 → 精修与合成(放大/超分、修手修脸、PS/AE 后期、抠图合成)→ 图生视频/文生视频(首尾帧、运镜、时长、一致性、对口型)→ 商业交付与迭代复现(参数留存、工作流文件复用、版权与可商用判断)」的完整创作-工程-交付决策链。覆盖 (a) 第一性张力 — **可控性/工程化复现(ControlNet/LoRA/ComfyUI 节点工作流/固定种子与参数/可复现可迭代)⇄ 随机性/抽卡玄学(出图随机、prompt 玄学、抽卡刷图、靠运气)**;**审美/创意/艺术指导主导(人的审美判断、概念设计、构图、镜头语言、艺术总监能力——'模型是地板,审美是天花板')⇄ 工具/模型/参数主导(堆模型能力、堆参数、堆 LoRA、以为换个大模型就行)**;**开源生态/本地可控(Stable Diffusion / ComfyUI / Flux / civitai / liblib 本地部署可定制可控可商用、显卡门槛)⇄ 闭源商业/易用黑盒(Midjourney / 可灵 / 即梦 / Runway 开箱即用但不可精控、按量付费、黑盒)**;**图像生成(T2I/I2I:Midjourney / SDXL / Flux)⇄ 视频生成(T2V/I2V:可灵 / Runway Gen-3 / Vidu / 即梦 / Sora,2024-2025 前沿从图迁移到视频)**;**生产力/商业量产(替代或增强传统设计/插画/电商/广告/影视分镜流程、降本提效、批量)⇄ 纯艺术创作/作者表达(个人风格、实验性、作品集)**;**效率/批量(一次出几十张筛选、商单走量、自动化 pipeline)⇄ 质量/精控(单张精修、商业可用级细节、修手修脸修穿帮)**。(b) 核心工作流 / pipeline(最标准、最易蒸高质量 + CLI 化)— 单图创作链:概念/参考收集(moodboard/风格定位/竞品)→ 模型工具选型(按可控性/风格/成本/可商用判断 MJ vs SD/Flux vs 国产)→ Prompt 与参数设计(主体/风格/光影/镜头/负向词/CFG/采样器/分辨率/种子)→ 首轮出图与方向筛选 → 可控迭代(ControlNet 控构图/姿态/线稿、LoRA 控风格/角色一致性、图生图重绘、inpainting 局部修)→ 放大超分与精修(高清放大/修手修脸/后期调色合成)→ 交付与参数留存(工作流文件/种子/参数复现)。视频创作链:分镜/脚本→首帧出图(图像链产出关键帧)→图生视频(运镜/时长/首尾帧/动态强度)→多镜头一致性(角色/场景/风格统一)→对口型/配音/配乐→剪辑合成→交付。ComfyUI 工程链:节点工作流搭建(加载器/采样/ControlNet/放大/面部修复节点)→ 参数化与复用→批量队列→工作流分享(civitai/openart workflow json)。(c) 工具栈 — 闭源图像(Midjourney v6/v7、DALL·E 3、Adobe Firefly、Ideogram、Recraft)、开源图像与底座(Stable Diffusion 1.5/SDXL/SD3、Flux.1 dev/schnell(Black Forest Labs)、ComfyUI(节点式)、Automatic1111/Forge WebUI、Fooocus)、可控生成插件(ControlNet、LoRA、IPAdapter、AnimateDiff、inpainting/outpainting、ReActor 换脸、面部修复 CodeFormer/GFPGAN、放大 ESRGAN/SUPIR/Topaz)、模型/资源社区(Hugging Face、Civitai、liblib 哩布、吐司 Tusi、openart)、视频生成(可灵 Kling、即梦 Dreamina/Jimeng、Runway Gen-3/Gen-4、Luma Dream Machine、Pika、Vidu、海螺 Hailuo、Sora、Wan 通义万相、MiniMax)、辅助(Magnific 放大、Krea、Photoshop/After Effects 后期、ChatGPT/Claude 写 prompt、reverse prompt 反推)。(d) 知识正典 — AIGC 创作 canon 横跨论文/工程/社区教程:奠基论文(DDPM、Latent Diffusion/Stable Diffusion(Rombach et al)、DiT、ControlNet(Zhang et al)、LoRA、DreamBooth、Textual Inversion、SDEdit、IP-Adapter、视频扩散 SVD/Sora 技术报告)、官方文档与模型卡(Stability AI、Black Forest Labs Flux、ComfyUI 文档、Midjourney docs、可灵/即梦官方教程、Hugging Face diffusers)、社区教程与工作流(ComfyUI 官方示例、civitai 文章、B站/YouTube 创作者长教程、openart workflow)、prompt 工程资源(Midjourney 风格库、提示词指南)。(e) figures/流派 — 模型/工具创造者(Midjourney David Holz、Stability AI Emad/Robin Rombach、Black Forest Labs(Flux,原 SD 团队)、ComfyUI comfyanonymous、ControlNet 张吕敏 Lvmin Zhang、AUTOMATIC1111、可灵/即梦/快手字节团队)、创作者 KOL 与教育者(Midjourney/SD 头部创作者、ComfyUI 工作流大神、AI 视频创作者、国内 B站/抖音 AIGC 教程作者、独立 AI 艺术家)、行业分析与评测(AI 工具评测、生成式 AI 创作生态观察者)。流派分歧:开源可控派(SD/ComfyUI/Flux 本地精控)vs 闭源易用派(MJ/可灵开箱即用)、工程化派(节点工作流/可复现/参数化)vs 玄学抽卡派(prompt 玄学/堆词/抽卡)、审美主导派('模型是地板审美是天花板')vs 工具主导派(堆模型堆参数)、图像派 vs 视频派、艺术创作派 vs 商业生产力派、国产工具派(可灵/即梦/liblib)vs 海外工具派(MJ/Flux/Runway)。(f) 行业话术/黑话 — 出图 / 抽卡 / 刷图 / 炼丹(训练 LoRA)/ 喂图 / 垫图(参考图/图生图)/ 控图 / 锁种子 seed / CFG / 采样器 sampler / 步数 steps / 重绘幅度 denoise / 提示词 prompt / 咒语 / 负向 negative / 权重 / tag / 大模型 checkpoint / 底模 / LoRA / ControlNet / 控线稿/openpose/depth / IPAdapter / inpainting 局部重绘 / outpainting 扩图 / 高清放大 / 超分 / 修手 / 崩手 / AI 味 / 风格 / 角色一致性 / 工作流 workflow / 节点 / 跑图 / 队列 / 文生图 T2I / 图生图 I2I / 文生视频 T2V / 图生视频 I2V / 首帧/尾帧 / 运镜 / 动态强度 / 时长 / 对口型 / 一致性 / 可商用 / 训练集 / 过拟合 / 翻车 / 穿帮。(g) 争议/批判 — 「版权与训练数据」(模型训练用未授权作品、生成图版权归属、可商用风险、洗图/伪原创)、「AI 替代创作者就业」(替代画师/插画师/设计师/摄影、行业冲击与转型)、「AI 味与审美趋同」(生成图同质化、套模板、缺乏原创审美)、「玄学 prompt 神话与割韭菜」('咒语'神秘化、'7天 AIGC 变现/月入过万'付费课割韭菜、抽卡当能力)、「工具快速迭代与淘汰焦虑」(模型月月换、工作流频繁失效、追新疲劳)、「开源 vs 闭源之争」(可控可商用 vs 易用黑盒)、「质量天花板」(修手修脸穿帮、长视频一致性、可控性仍有限)。(h) 大量隐性创作手艺 + 审美 + 工程 + 沟通软技能(prompt 的描述功底与审美词汇、参数与采样的手感、ControlNet/LoRA 的组合控制经验、节点工作流的搭建与调试、批量筛选的审美眼力、修图合成的后期功底、镜头语言与运镜设计、角色/风格一致性的控制、商业需求的翻译与交付、版权与可商用的判断)是高 tacit、靠大量实操跑图 + 审美积累 + 工程调试经验的核心。水分极高(大量'咒语速成/一键变现/AI 绘画割韭菜'营销

日本語の概要は準備中です。原文の説明を表示しています。

swaylq/master-skill1492026年9月6日 更新

swaylq のスキルをすべて見る

このスキルの問題を報告する