@wji.ai/cli 0.4.0 → 0.5.1

package/skill/short-drama/reference/story-cultivation.md

@@ -0,0 +1,109 @@
+# Story genre `cultivation` — 玄幻 / 修仙 / 系统流
+
+修仙 · 玄幻 · 系统 / 金手指 · 都市异能 · 废柴逆袭 · 神医 / 鉴宝 系统. The level-up genre. Pairs with the
+`costume-idol` look (修仙) or `scifi-soft` (都市异能 / 系统), and with the `wuxia-fantasy` or
+`superhero` action style. Shared machinery: `story.md`.
+
+## 1. The promise — visible growth
+
+- **The engine is the upgrade loop**: a nobody climbs a fixed ladder, step by step. Each trope pays a
+  feeling: 废柴逆袭 ("someday I'll count"), 退婚打脸 (trampled dignity repaid), 捡漏得宝 (unearned luck),
+  越级杀敌 (breaking the rules).
+- **扮猪吃虎 is an information-gap pleasure**: the hero knows more (time loops, texts from the future,
+  the system).
+- **系统 suits 短剧**: an instant-payoff 金手指 that pairs with any hot theme (荒年 + 系统 passed a billion
+  plays). 男频 has turned toward quality: the grassroots hero growing up, the thrill of controlling
+  power.
+- **The 女频 version differs**: 满级大佬 (starts at max), 团宠, 升级流 — the biggest 女频修真 hit starts her
+  as a goddess who clears villains with a wave. So the genre has a **`startPower`**: `废柴` (climbs),
+  `隐藏大佬` or `满级` (hides) — and the rules change with it.
+- **The market**: 红果 pushed 男频 in 2025, naming 玄幻仙侠 and 都市异能; on AI 漫剧 charts 玄幻仙侠 leads with
+  废柴逆袭 + 升级打怪 + 系统面板.
+
+## 2. The series — a realm ladder and map changes
+
+- **The ladder**: 练气 → 筑基 → 金丹 → 元婴 → 化神 → 炼虚 → 合体 → 大乘 (→ 渡劫), each with 初期 / 中期 / 后期 / 大圆满.
+  Subdivide so small highs come regularly, layered with big ones — but in 短剧 **pick 5–7 realms and
+  climb 3–5 on screen**.
+- **换地图** is levelling up made visible (village → county → sect → capital → upper realm); each map's
+  enemies, resources and ranks climb in step. The loop: beat stronger enemies → gain a skill or a
+  treasure → clear the region → move on.
+- **On a 60–100 episode run**: golden (E1–10) humiliation → 觉醒 / binding → first gain → first 打脸 →
+  first breakthrough; 2–4 map arcs of 15–30 episodes, each with a rival and a boss realm, a major
+  breakthrough ending each arc; **the peak held for the last 10–15 %**. A tangible gain every 3–5
+  episodes, a major breakthrough every 12–20.
+- **Against power collapse (战力崩坏)**: fixed multipliers between realms, few top-tier fighters, seals,
+  political duties for the powerful, equivalent-exchange economies — and keep the story short.
+  Progression-fantasy craft: power should *buy harder problems*; bottlenecks, not grinding.
+
+**Tracked in the outline** (gate `story-cultivation`): `outline.powerSystem: { realms: [{ id, name,
+visual }] }` (low → high), `outline.goldenFinger: { type, name, activationEp, rules[], costs[] }` (one
+or two), `outline.startPower`, `outline.upgrades[]: { ep, char, to, cause, cost }`,
+`outline.fights[]: { ep, foe, heroRealm, foeRealm, foeRevealedEp, edge? }`.
+
+| rule | default | why |
+| --- | --- | --- |
+| `golden-finger` | activated by E2, with rules and costs; at most two | the golden opening; limits matter more than powers; a sprawl of cheats bores |
+| `upgrade-cadence` (`废柴`) | no 12-episode stretch without a breakthrough; one in the golden phase | the level-up is the promise |
+| `no-level-skip` (`废柴`) | one realm at a time; no fall back unless it is a recorded cost | ladder integrity |
+| `cause-and-cost` | every breakthrough has a cause and a cost (time, resources, injury, exposure, a debt to the system) | Sanderson: limits > powers |
+| `foe-known` | the foe's realm is known before the fight; at most one realm up, and then with an edge set up earlier | makes 越级 legible; no deus-ex-system |
+| `peak-held` (`废柴`) | the top realm reached in the last 15 % | keeps the climax |
+| `hidden-master-reveals` (`隐藏大佬` / `满级`) | a new layer of 身份揭露 at least every 10 episodes | the tension moves to the hidden identity |
+
+## 3. The episode
+
+- The micro-arc fits one episode: conflict → the system activates → crushing power.
+- **Every episode moves a ledger quantity** (a resource, a skill, an item, reputation or a realm) — not
+  a full level each time.
+- **System notifications are ready-made cliffs**: 【叮！任务发布：…失败惩罚：…】 — a concrete task, a
+  reward that builds anticipation, a penalty grasped at once. Task or penalty → 危机; the unopened
+  reward box → 道具; "breakthrough in progress…" → 结果; the enemy's realm revealed → 身份 / 反转. At
+  most two system hooks in a row.
+- **Breakthrough cliffs**: cut on the aura surge or the tribulation cloud; pay off in the next
+  episode's first 15 s with an onlooker's shock ("筑基了?!").
+
+## 4. Characters
+
+- **The 废柴 hero**: a visible humiliation (退婚, expelled from the sect) and a goal beyond power
+  (family, revenge, someone to protect).
+- **The system or the 老爷爷** has a personality, blind spots, limits, even an agenda — never
+  omniscient in practice. One or two 金手指 for the whole story.
+- **A rival per realm or map**, within about one realm of the hero — the revenge ladder indexed by
+  realm (`story-revenge.md`).
+- **The love interest** has her own stake; the 退婚 fiancée and the loyal girl are worn out.
+- **For 满级 heroes**, the tension is hiding, protecting others and consequences, not whether she wins.
+
+## 5. Beats and freshness
+
+觉醒 / 绑定系统, 系统提示, 退婚, 宗门大比, 越级挑战, 秘境, 丹药 / 法宝, 拍卖会, 天劫, the 扮猪吃虎 reveal, 下山即无敌.
+Fresh through **融梗** (a 无情道 system in 女频修真, famine + stockpiling, the time loop, texts from your
+future self, 签到, 全民转职, 规则怪谈) and through conviction — "make the unreasonable believable".
+
+## 6. Dialogue and AI video
+
+The system speaks in bracketed UI copy: short, numeric, deadpan. **Onlookers say the realm aloud**
+("他竟是金丹后期!") — exposition and payoff in one line. Contempt lines (区区练气) set up the 打脸; understate
+the hidden master. **For AI video**: give each realm one fixed visual signature (an aura colour, a
+particle, an eye glow) — `powerSystem.realms[].visual`; ration full-screen set pieces (天劫, armies)
+for breakthroughs and the climax; prefer glows, runes, UI panels and flying swords over crowds and
+long choreography.
+
+## 7. What fails, and regulation
+
+Power creep from open-ended multipliers; a sprawl of 金手指; rules the story does not follow; the
+deus-ex-system (a new ability exactly when needed — Sanderson's First Law); the same plot shot a
+hundred times. **Regulation**: the micro-drama measures in force from 2026-09-01 ban promoting
+superstition and cults and glamorising violence, and require an AI label on every episode; compliance
+guides treat 招魂, 鬼上身, 养小鬼, 冥婚, and fortune-telling or exorcism shown as real as red lines. **An
+invented world is the safe setting**; a modern-city 神医 / 风水 / 鉴宝 is the risky one — never show
+talismans or feng-shui working as real cures or luck, never disparage medicine (`story-advice` flags
+it on the mainland).
+
+## 8. Sources (read as search summaries)
+
+CSDN on 修仙 writing and 金手指 design; 知乎 on 金手指, 扮猪吃虎 / 越级 and power-system collapse; 澎湃 网文论 (the
+MMORPG loop) and 女主修真赛道; DataEye 2025; 钛媒体 on 系统 hits; Sudowrite and Writer's Daily Practice on
+progression fantasy; Royal Road on pacing; Brandon Sanderson's laws; 百度百科 on 升级流 and 系统; 知乎 / 腾讯 on
+融梗 and 信息差; 腾讯 on 男频 recovery; 搜狐 and 蓝鲸 on 漫剧 charts; 起点 on realm ladders; 知乎 on homogeneity; 新华网,
+NRTA and 17173 on the 2026 measures; Byerisk on superstition red lines; 36氪 and 新京报 on AI 仙侠 costs.

package/skill/short-drama/reference/story-family.md

@@ -0,0 +1,119 @@
+# Story genre `family` — 家庭伦理
+
+婆媳 · 豪门恩怨 · 萌宝 · 亲情 · 认亲 / 寻亲 · 养老 · 家庭复仇 · 年代家庭. The genre of 《家里家外》 and
+《十八岁太奶奶驾到》, and of the fastest-growing audience in 短剧. Shared machinery: `story.md`.
+
+## 1. The promise — being needed, being loved
+
+- **Who watches**: in 2024 viewers aged 40–59 were 37 % of the short-drama audience and 60+ were 12 %
+  and rising fast; 中老年 titles take close to half of some monthly heat charts, and older men are a
+  large share of payers. Hits like 《家里家外》 and 《我在八零年代当后妈》 also reach students; 原生家庭
+  stories (《好一个乖乖女》) pull in younger women. **Serve both poles**: the parent (the wronged
+  mother, the unfilial child) and the child (the 原生家庭 wound, the found family).
+- **The underlying need is 被需要 and 被爱.** The older viewer's first pain is their children: the
+  mother who sold her flat for her son's wedding home. The payoff is the ungrateful child finally
+  seeing the sacrifice — and a double happy ending, love and family.
+- **The shift is 从强刺激到强共鸣**: 情怀, 家和亲情, "家才是顶流剧本"; family as the unit, not the company or
+  the couple.
+- **Four emotions**: 共鸣 (that's my mother-in-law), 泪点 (sacrifice seen too late), 家的温暖 (a found or
+  blended family), 扬眉吐气 (the wronged one vindicated before everyone who looked down on her). The
+  last overlaps with `revenge` — but the target is kin, so the end is 和解 or 清醒的切割, not destruction.
+
+## 2. The series — a family web that reorganises
+
+- **Pressure runs long**: 中老年 titles hold the heroine down for twenty or thirty episodes before the
+  big payoff, much longer than 爽剧 — so give small wins early (a reason to stay) and the full public
+  vindication in the middle third, then a larger one in the climax.
+- **A web, not a line**: three lines work — the present line drives the plot, a memory line explains
+  the past, a hidden line plants motives, with an emotional anchor every few episodes. Pivotal
+  revelations happen in **ensemble scenes where everyone is present**.
+- **Family as a system**: the 偏心 parent runs the golden-child / scapegoat engine; the 老实人 son or
+  husband is the enabler; the 萌宝 is the mascot and truth-teller; the wronged daughter-in-law is the
+  scapegoat who breaks the role. The series is the system reorganising — so **every member ends
+  somewhere**.
+- **Secrets**: tell the audience the blood secret early (信息差, by E5); reveal it character by
+  character (the intercepted DNA result, the 信物, the witness); stage the full public 认亲 at a
+  gathering at 55–85 %; leave episodes for the fallout.
+- **爽 and 泪 alternate**: humiliation → small 爽 → a warm or tearful beat → bigger humiliation →
+  bigger 爽. Never more than about three 爽 episodes without a tear — that is what separates family
+  drama from revenge.
+
+**Tracked in the outline** (gate `story-family`): `characters[].family: { generation, role,
+minor?, endState: 和解 / 清醒切割 / 惩戒 / 离世 }`, `outline.secrets[]: { id, content, audienceKnowsEp,
+plantEp, hintEps[], revealEp, revealTo }`, `outline.vindications[]: { char, ep, venue, witnesses[] }`,
+`outline.reconciliations[]: { pair, woundEp, triggerEp, turnEp, endState }`.
+
+| rule | default | why |
+| --- | --- | --- |
+| `family-web` | ≥ 5 family members over ≥ 2 generations | the genre is a system, not a duel |
+| `every-member-resolves` | every lead and support family member has an end state | the system visibly reorganises |
+| `tear-cadence` | ≥ 20 % 情感释放 episodes, never more than 6 in a row without one | 泪 is the promise |
+| `public-vindication` | a vindication with ≥ 3 family witnesses at 25–70 %, a larger one in the climax | "压到二三十集", then before everyone |
+| `secret-hygiene` | every secret planted before it is revealed with ≥ 2 hints between; the main secret known to the audience by E5, revealed at 50–85 % | 信息差; room for the fallout |
+| `earned-reconciliation` | wound < trigger (evidence of sacrifice) ≤ turn; ≥ 10 episodes from wound to turn | repentance earned by sacrifice, not by her wealth (富贵门前皆孝子) |
+| `child-compliance` | a minor never lands an 打脸 / 复仇 / 智商碾压 | NRTA 2026 儿童类微短剧 notice: no adultification, no 啃小, no children as tools |
+
+## 3. The episode
+
+- **Cold open on a concrete family wound**: the mother eating leftovers while the in-laws feast, the
+  soup knocked away, the old woman outside her own house. The wound, not the violence — takedowns
+  cite 刻意渲染.
+- **The tear beat in the back half**, usually a memento (the old photo, the handmade shoes, the
+  存折, the medical report) or a silent sacrifice seen by a third party; cut on it (情感) or undercut it
+  with a threat (危机).
+- **Cliffs**: the DNA envelope unopened (道具 / 身份); the elder collapses, a diagnosis (危机); the
+  unfilial child about to sign away the house (结果); the 信物 seen by the wrong person (道具); the 萌宝
+  blurts "you're my daddy" (情感 / 身份); the banquet door opens (身份 / 反转).
+- **End a vindication on the key relative's face** — shame and regret. The villain's comeuppance is
+  not enough; the repentance face is what this audience pays for.
+
+## 4. Characters
+
+- **The wronged mother / daughter-in-law / elder**: 为了孩子, 委屈求全, and hidden value (a status, a
+  skill, a past sacrifice). Give her the moment of choosing herself — 半生为母，余生为己. Pure martyrdom
+  is dated. (The 2024 婆婆爽剧 wave flipped it: the mother-in-law as the protagonist.)
+- **The 萌宝**: builds arcs (the cold father warms), drives 认亲, fulfils fantasy — as an **innocent
+  truth-teller or bridge** who notices, asks or carries a 信物. Never a schemer, a 霸总 or a bully; no
+  adult lines.
+- **The unfilial child, the 偏心 parent, the 白眼狼**: they repent — earned by evidence of sacrifice.
+- **The elder** is the arbiter at the 寿宴 and the holder of the will.
+- **Antagonists stay human**: a real motive (fear of poverty, an inherited 偏心, protecting their own
+  child), one scene of their own pain, a proportionate consequence (loss of standing, not violence).
+  《家里家外》 won by treating women's relationships warmly instead of 斗小三.
+- **Ensemble**: every core member wants something; rotate the focus; gather everyone for the
+  revelations.
+
+## 5. Beats and freshness
+
+Staples: 认亲 / 寻亲, 真假千金, 抱错, the DNA test, the 寿宴 / 满月宴 / 订婚宴 humiliation and reveal, 分家产 /
+遗嘱, selling the house for the son, being thrown out, 萌宝助攻, 闪婚老伴, 黄昏恋, 重组家庭, 后妈, 年代. Fresh
+through **period and dialect specificity** (1981 Sichuan, the women's volleyball team, 天府可乐),
+**lived detail** (the writers interviewed hundreds of locals), **生活流 over 套路** (代际和解, 国企改制),
+**inversions** (the good stepmother, the victim mother-in-law, the 18-year-old great-grandmother), and
+**series-ability** (more of what viewers loved). Plan at least two whole-family gatherings (寿宴,
+年夜饭, 满月, 婚宴, 葬礼, 分家) — one in escalation, one in the climax.
+
+## 6. Dialogue
+
+**Dialect** is a selling point (《家里家外》, the first all-Sichuanese short drama). **Plain language**;
+each character's styling fits their life. **No 说教**: put the moral in an object or an act (the
+存折, the unworn new clothes, the mother walking in the rain); one quotable line per arc is enough.
+
+## 7. What fails, and regulation
+
+Homogeneous 婆媳 slap-fests; caricatured women; money-worship endings; logic sacrificed for emotion;
+repentance triggered by wealth; relatives left unresolved; a 萌宝 as a plot vending machine.
+**Regulation**: the NRTA 2025-02 notice; takedowns for 刻意夸大渲染婆媳矛盾 and violence; the 2026-01
+儿童类微短剧 notice (反成人化, 防啃小, 防工具化); the June 2026 campaign (harm to minors, wealth worship,
+distorted marriage values, violence and revenge). Keep at least two hierarchical pairs (婆媳, 母子,
+父女…), not 婆媳 alone, and end every abuse resolved.
+
+## 8. Sources (read as search summaries)
+
+人民日报 on 中老年微短剧 (audience shares); 36氪 / AgeClub "被需要与被爱"; 虎嗅 on how 中老年短剧 跑通; 36氪 on
+霸道总裁俏阿姨 and 婆婆爽剧; 21经济网 on 银发经济 and payers; 澎湃, 新华网, 每经 and 界面 on 《家里家外》; 网易 /
+澎湃 on 听花岛 (《太奶奶》); 搜狐 on 《太奶奶》; 虎嗅 on 年代短剧 生活流 (2026); 中国作家网 on 《好一个乖乖女》; 新浪 on
+《我在八零年代当后妈》; 搜狐 / DataEye, 网易 and 央广网 on 萌宝短剧; 新华网 on the 儿童类 notice (2026-01) and the 2026
+专项治理; NRTA 2025-02 notice; 新华报业 on takedowns; 光明网 on distorted family images; 流媒体网 on 卡点; 新浪财经
+on the three-line structure; 知乎 on 真假千金; No Film School and Writers Helping Writers on ensembles;
+Simply Psychology on family-system roles.

package/skill/short-drama/reference/story-horror.md

@@ -0,0 +1,124 @@
+# Story genre `horror` — 恐怖 / 灵异 / 惊悚
+
+心理恐怖 · 规则怪谈 · 民俗恐怖 · 都市怪谈 · 怪兽 · 灵异. It builds on `mystery` (questions and clues) and adds
+the threat, its rules and the dread. **The market decides what is allowed** (`params.market`,
+§7). Shared machinery: `story.md`.
+
+## 1. The promise — safe fear
+
+- **Recreational fear**: fear enjoyed because the frame is safe — a flight simulator for anxiety
+  (adrenaline junkies, white-knucklers, dark copers). The genre promises rising fear, then release:
+  you survived, and you understood.
+- **King's ladder**: terror (what you imagine — the best) > horror (seeing something wrong) >
+  gross-out (the lowest, and effectively banned on the mainland). Live at terror and horror.
+- **Dread is an information gap in the audience's favour** (Hitchcock's bomb): mystery withholds *what
+  happened*; horror shows *that something is coming* and withholds when, where and what.
+- **Curiosity about the rules** (规则怪谈): the viewer rebuilds the hidden world from contradictory
+  rules (细思恐极) — in 短剧 an 智商碾压 payoff when the lead decodes the true rule.
+- **The market**: overseas, thriller and horror are reported to grow fast among 18–30 viewers
+  (trade-blog figures); psychological, domestic and paranoia thrillers suit 9:16 better than chases.
+
+## 2. The series
+
+- **Knowledge drives the acts**: unaware → suspecting → certain. The golden segment establishes
+  normality and puts **the first sign of wrongness in E1, within 15 s**, and one threat event (a victim,
+  a disappearance, a near miss). Escalation raises the stakes in stages (property → a secondary
+  character → a loved one → the lead). **The midpoint** is usually the rule or nature reveal, often
+  after a false win — the lead stops reacting and starts driving. Late escalation shows the monster
+  more (the Jaws rule: hold the full reveal as long as possible); allies fall; the "safe" rule turns out
+  corrupted. **The climax is won by exploiting the rules**, not force.
+- **规则怪谈 anatomy**: the rulebook is the setting (a hospital, a school, a hotel, a compound, a zoo);
+  a human side writes rules to protect, an abnormal side contaminates them; contradictions (the
+  student vs the teacher handbook; red ink) force the characters to tell true rules from false. Beat
+  map: enter the instance and receive the rulebook → the first rule-breaker dies (the rules are real)
+  → the lead spots an anomaly → a contradiction exposes a false rule (midpoint) → the hidden true rule →
+  the rules turned against the monster → the way out, and a sting.
+- **Endings**: overseas and J-horror allow a sting (the curse spreads); **mainland endings close
+  rationally with a positive value** (《蛇影谜踪》 closes on 敬畏自然).
+
+**Tracked in the outline** (gate `story-horror`): `outline.horror: { subgenre, threat: { nature,
+firstSignEp, glimpseEps[], revealEp, explanation: { type: 人为 / 幻觉 / 科学 / 骗局 / 精神 / 超自然, revealEp,
+clueEps[] } } }`, `outline.horrorRules[]: { id, text, isTrue, shownEp, brokenEp?, consequence }`, and on
+each outline episode `scares: [氛围 / 假警报 / 惊吓 / 揭示 / 规则惩罚]`.
+
+| rule | default | why |
+| --- | --- | --- |
+| `scare-cadence` | a real scare (惊吓 / 揭示 / 规则惩罚) in ≥ 80 % of episodes | "a cliffhanger every minute"; the core promise |
+| `false-alarms` | a false alarm is followed by a real scare in the same episode, never two episodes running | the cat scare sets up the real one |
+| `early-threat` | the first sign in E1 | the 15-second hook |
+| `hide-the-monster` | the full reveal past the half, with ≥ 2 glimpses before it | the unseen is scarier; the Jaws rule |
+| `fair-rules` | a rule is shown before anyone is punished by it | dread needs the audience to know the rule |
+| `rule-horror` (规则怪谈) | ≥ 5 rules, the first by E2, a false or contradicting rule by the midpoint | true vs contaminated rules is the game |
+| `mainland-explanation` (`market: mainland`) | the threat is explained by 人为 / 幻觉 / 科学 / 骗局 / 精神, revealed in the climax, with ≥ 2 clues planted ≥ 2 episodes before | ghosts confirmed real, fortune-telling that comes true and 封建迷信 are red lines; an unplanted "it was a dream" cheats |
+
+## 3. The episode
+
+- **The scare cycle**: quiet → build (sound drops, focus narrows) → a false alarm → relief → the real
+  scare, often from a misdirected angle. Ration the false alarms.
+- **The 60-second template** (TikTok horror): 0–3 s a hook (a sound, an image, a line of text); set up a
+  pattern and crack it a little each pass; the turn; a lingering image — the cliff.
+- **Imply, don't show**: "let the mind supply the rest" — Blumhouse avoids CGI because the imagined is
+  scarier; this is also what makes horror cheap for AI video.
+- **Cliffs** are mostly 危机 (the threat imminent) and 反转 (the ally is the monster; the rule was false).
+- **Vary the mechanism**: 《逃出绝命岛》 was criticised for repeating "狼来了" (she warns, nobody listens,
+  someone dies) until it bored. Mark the escalation with an `intensity` per episode if you track it; put a
+  quieter breather after a peak.
+
+## 4. Characters
+
+- **The survivor (final girl)**: watchful, level-headed, the first to sense something amiss and the
+  only one to piece the pattern together — in 短剧 each 爽点 is often her reading the rules right.
+- **The skeptic** denies the threat and dies or converts; on the mainland the skeptic can be *right
+  about the mechanism* (a human is behind it) while wrong about the danger.
+- **The expert** (the folklorist, the old caretaker, the rules veteran) delivers the backstory and is
+  lost by the late escalation so the lead acts alone.
+- **The first victim** proves the threat and the rules, by E3.
+- **The human antagonist** (the snake dealer, the one faking the haunting) — required on the mainland,
+  useful everywhere for 打脸 / 复仇.
+- **No idiot plot**: every bad choice has a motive — forced, deceived, protecting someone, a false rule
+  believed in good faith.
+
+## 5. Beats and freshness
+
+The rulebook reveal; the first rule-breaker's death; "don't look back / don't answer"; the wrong
+mirror or door; the count off by one; the ally who knows too much; the false safe zone; the red-ink
+rule; the monster at the frame edge; the curse spreading like information (J-horror); the wronged
+dead with a grievance (onryō, the K-horror family trauma). **Fresh**: 民俗 / 志怪 anthologies (the pawnshop
+that takes cursed objects), creature features (《逃出绝命岛》, 《蛇影谜踪》, hundreds of millions of plays),
+规则怪谈 / 诡异 / 末世 漫剧, and blends (horror + family, + 探案, + comedy). **Mainland-forbidden or high-risk**:
+ghosts confirmed real, fortune-telling or feng-shui that works, karma as a real mechanism, detailed
+religious or cult ritual, gore and "画风观感不适", minors in terror scenes, pure ghost horror in AI 漫剧.
+
+## 6. Dialogue and tone
+
+Short, plain, concrete. **Rules in a flat bureaucratic register** — the authority is where the fear
+comes from ("even nonsense sounds reasonable written as a rule"). Understatement over exposition:
+silence, a sound cue, one wrong detail; characters do not narrate their fear. A dry line after a false
+alarm resets the baseline so the next build lands harder. On the mainland: 爽而有度, fear resolved into
+reason, a closing value (反迷信, 敬畏自然, 惩恶).
+
+## 7. The market — mainland and overseas
+
+- **Mainland** (`params.market: mainland`, the default): the 2017 content review code bans promoting
+  superstition and rendering terror; the absolute red lines are 色情低俗, 血腥暴力恐怖, 封建迷信 and 错误价值观;
+  tiered review and 先审后播, and since 2026 platforms review every uploaded micro-drama. **The practical
+  line**: 灵异 as atmosphere is allowed **when the ending explains it** (hallucination, fraud, a human,
+  the mind); never confirm ghosts, never let fortune-telling come true. The "建国后不许成精" rule is a myth,
+  but the pattern is real; period and 聊斋-style settings get more leeway. The 红果 / 抖音 漫剧 rules (from
+  2026-04-07) ban 风险迷信 and 画风观感不适 and reportedly pure ghost horror and supernatural 金手指 for AI
+  漫剧; the viral "three banned genres" list for live action was officially denied — treat 灵异 as
+  high-risk, not banned.
+- **Overseas** (`params.market: overseas`): the supernatural may be a real mechanism (cursed manors,
+  werewolves, onryō curses); the limits are app-store and ad policies on gore and minors; keep the
+  threat's rules consistent and answer the central question even if the ending stings.
+
+## 8. Sources (read as search summaries)
+
+Wikipedia on 红果 and 规则怪谈; 知乎 and 腾讯频道 on 规则怪谈 craft; 萌娘百科 on 规则类怪谈 and 建国后不许成精; 中国作家网 on
+规则怪谈 and control; Filmustage on vertical genres (trade figures); 新浪财经 on 2025 漫剧; Byerisk on superstition
+red lines and the red-line table; IT之家 on 红果's denial; 36氪 / 腾讯 on 《蛇影谜踪》 and 《逃出绝命岛》; Far Out
+(Hitchcock); No Film School on Jaws and Blumhouse; Wikipedia on horror and terror and the final girl;
+Psychology Today on recreational fear; Kuciak on horror's three acts; Stone Age Herbalist on J-horror;
+Backstage on jump scares; TIME and Little House of Horrors on TikTok horror; Filmora on ReelShort; Asian
+Movie Pulse on K-horror; 腾讯 红榜 on 民俗志怪; 腾讯 on the 红果 漫剧 rules; 界面 on the decline of Chinese horror
+film; 新华网 on the 2017 review code, the 2025 tiered review and 凡播必审 (2026).

package/skill/short-drama/reference/story-mystery.md

@@ -0,0 +1,127 @@
+# Story genre `mystery` — 悬疑
+
+推理 · 刑侦 · 悬疑复仇 · 惊悚. In 短剧 the pure puzzle is rare and hard to sell (红果's 悬疑推理
+category drew 200+ titles, very few past 5,000万 heat); the hits **blend suspense with an emotional
+payoff** — 《引她入室》 is a family revenge story in a suspense frame. So the default here is
+悬疑 + 复仇 / 身份 爽点, with clue-solving as the way the 智商碾压 or the 身份揭露 is delivered.
+Shared machinery: `story.md`.
+
+## 1. The promise — curiosity, dread, the click
+
+- **Three pleasures**: curiosity (what happened, who did it), dread (what will happen, when), and
+  the click — the answer was in plain view all along.
+- **Who knows what** (McKee): mystery (the characters know more → curiosity), suspense (both know
+  the same → worry), dramatic irony (the audience knows more → dread).
+- **Hitchcock's bomb under the table**: surprise buys 15 seconds; showing the bomb and the clock
+  buys 15 minutes. **Keep the central whodunit a mystery, but run most episodes as suspense or
+  irony** — the audience sees the threat and the clock, the character does not. Irony delivers
+  tension in 60 seconds; a pure puzzle needs setup time the format does not have.
+
+## 2. The series — one question, layered secrets
+
+- **Play the cards early.** Plant every 梗 and 伏笔 in roughly the first 30 episodes and use the
+  rest to decode them. The first ten are the golden window: the main line, the core hook, and
+  "what kind of story this is".
+- **A central question over everything, small questions that last an episode or two** ("a ball of
+  questions rolling"). Each answer exposes a deeper question.
+- **逻辑闭环**: every step from the crime to the solution is explained, and every detail the
+  investigator uses was set up earlier.
+- **Case-of-the-week plus a through-line** works (《朱雀堂》: each case closes, the main-line
+  伏笔 accumulate); so does a jigsaw of timelines (《漫长的季节》, the two lines merging at the end of
+  episode 1 to pay off the device early).
+
+| phase | what |
+| --- | --- |
+| golden (~E1–10) | the inciting crime or secret in E1; the **central question stated by E3**; the investigator, the victim and all major suspects — **the culprit included** — on screen; a first small question answered by E3; a layer-2 reversal at the paywall |
+| escalation | layered secrets; false suspects rise and fall; a **midpoint twist at 40–60 %** that reframes known clues rather than adding new ones |
+| climax | the central truth, **assembled from planted clues**, at 80 % or later; the justice or revenge payoff; 1–3 episodes of aftermath |
+
+**Tracked in the outline** (gate `story-mystery`):
+- `outline.mysteryMode`: `whodunit` (default) · `inverted` (the audience knows the culprit — the
+  question is how they will be exposed) · `thriller` (a threat, not a puzzle: no culprit rule).
+- `outline.questions[]: { id, layer: 1 central / 2 major / 3 local, question, openEp, answerEp, clues[] }`.
+- `outline.clues[]: { id, kind: true / red-herring, what, plantEp, prop?, echoEps[], dismissedEp? }`.
+- `characters[].mysteryRole` (`investigator` `culprit` `accomplice` `suspect` `unreliable-ally`
+  `victim` `witness`) and `firstEp`.
+
+**The rules** (researched defaults, waivable on the user's request):
+
+| rule | default | why |
+| --- | --- | --- |
+| `central-question` | exactly one layer-1 question, opened by E3, answered at ≥ 80 % (runs ≥ 10) | the golden window; the truth in the climax |
+| `culprit-early` | exactly one culprit, first on screen by E5 / the golden phase (never after 30 %) | Knox 1, Van Dine: the culprit appears early and plays a real part |
+| `clue-count` | an answered question has ≥ 3 true clues (layer 1), ≥ 2 (layer 2), ≥ 1 (layer 3) | fair play: every clue to the solution is shown first |
+| `clue-lead` | a clue is planted ≥ 1 episode before a local answer, ≥ 3 before a major one, ≥ max(10, 15 %) before the central one; one central clue in the golden phase | too close and they see it coming; too far and they forget |
+| `clue-echo` | a clue planted more than 20 episodes before its payoff is echoed within 15 episodes of it | fragmented viewers need a reminder |
+| `red-herrings` | fewer red herrings than true clues; each dismissed before the truth | too many and the audience feels tricked, not tested |
+| `open-questions` | at most 6 open at any episode | the fragment viewer is lost past that |
+| `clue-on-screen` (script) | the beat that plants a clue is marked `clue: "K1"` in its plant episode | a clue is something the camera shows, never narration |
+
+`story-advice` also notes a whodunit with fewer than 3 or more than 5 viable suspects and 反转
+cliffs above 35 % (twist fatigue).
+
+## 3. The episode
+
+- Conflict or a reversal within 15 s, the story moving by 30 s, a suspense beat in the last 10 s;
+  95 of 100 endings leave a hook. 红果 asks for three hooks a minute in the first 30 % of a 拉新
+  drama.
+- **Answer one, open one or two.** Each episode closes or partly answers a small question and
+  opens or escalates another. Quality over quantity; alternate 紧张 and 松口气.
+- **Hooks** (红果 tutorial): an unfinished accusation; an identity hint ("你真以为我不知道你是
+  谁？"); a deadline; the cut at the instant emotion erupts; an unexpected figure; **a familiar face
+  flashing across a CCTV frame**.
+- **Cliffs** by hook type: 身份 (the anonymous helper is…), 危机 (the bomb clock), 道具 (a clue object
+  where it should not be), 反转 (an accepted fact overturned), 结果 (the body, the test result),
+  情感 (the ally is lying).
+
+## 4. Characters and suspicion
+
+- **The investigator** has a distinctive skill or persona and solves the case through their own
+  reasoning on shown clues; in 悬疑复仇 it is often the wronged protagonist.
+- **The culprit** appears early, is someone the audience cares about, is not obvious, and their
+  thoughts are **never shown** (no inner voice from the culprit before the truth). One mastermind;
+  helpers are accomplices.
+- **3–5 viable suspects**, each with motive, means and opportunity on screen. Rotate the prime
+  suspect ("前一刻认定的凶手，下一刻可能因新证据被推翻"). Give the real culprit one fair moment of
+  suspicion, then clear them with an explanation that is plausible and false.
+- **The unreliable ally** — the helper with a hidden agenda — is the standard 反转 engine.
+
+## 5. Clue craft
+
+- **Fair play**: every clue leading to the solution is shown before the reveal; the investigator
+  uses nothing the audience has not seen; no supernatural agency or unknown poisons unless the
+  genre says so (Knox, Van Dine).
+- **Chekhov**: every prop, skill or line introduced is a promise — pay it off or cut it.
+- **Hide in plain sight**: plant the true clue inside a beat that is doing something else (a 爽点
+  slap, an argument), in the background of a wide shot, or with an innocent explanation (the ring's
+  "flaw" read as marital symbolism). Bury it fourth in a list. **Red herrings may get the
+  close-up; true clues usually should not at the plant.**
+- **Pay off with a close-up and a flashback insert**, so fragmented viewers reconnect it. On the
+  storyboard the plant is often an `evidence` cut in a wide, the payoff an `evidence` close-up.
+- **Name puns** (吴池 = 无耻, 郑萋 = 正妻) are cheap, legible foreshadowing that suits 短剧.
+
+## 6. Dialogue and tone
+
+Separate what each character **knows** from what they can **prove**; every interrogation question
+is a thrust, not a static Q&A. Subtext is want or fear expressed indirectly, carried by the body.
+**悬疑创作是克制的艺术**: concrete details, no overflowing emotion. Characters withhold for a
+reason of their own, never because the writer needs a delay; every lie told on screen can later
+be caught against something already shown. "笑着去讲悲伤的故事" hits harder than heavy lines.
+
+## 7. What fails
+
+强行反转 and 降智 characters; twists for show; the mystery box that never pays off; 高开低走 (a big
+opening that collapses); a clue the audience never saw, a late culprit, a supernatural solution;
+too many red herrings; relentless intensity with no character growth; too many characters too
+early.
+
+## 8. Sources (read as search summaries)
+
+腾讯 / 搜狐 on 逻辑闭环 and clue density; 新浪 (producer 陈璐); 流媒体网 on the 红果 悬疑 category, 古装
+悬疑 craft (《朱雀堂》) and 卡点; 腾讯 / CSDN "all cards in the first 30 episodes"; 红果 tutorial 06
+(hooks) and 编剧第一课 (3 hooks a minute); 搜狐 (张静思); 电影界 / 百度百科 on 《引她入室》; 网易 / 知乎 on
+《漫长的季节》; 界面 (潘依然, 《隐秘的角落》); 知乎 (点灯) on restraint and 高开低走; DataEye 2025; 澎湃 on
+红果 2025; Hitchcock / Truffaut and Bordwell on the bomb; Knox's commandments; Van Dine's rules;
+McKee on mystery vs suspense; StudioBinder and iwrity on Chekhov's gun; Writer's Digest and The
+Write Practice on clues and red herrings; nailyournovel on the ball of questions; InkTip and
+Filmustage on vertical drama; interrogation craft (StudioBinder, Bang2write); 文汇报 on 全程高能.

package/skill/short-drama/reference/story-palace.md

@@ -0,0 +1,103 @@
+# Story genre `palace` — 宫斗 / 权谋
+
+宫斗 · 宅斗 · 嫡庶之争 · 夺嫡 · 权谋复仇 · 女帝 / 女相 大女主. It pairs with the `costume-dark` or
+`costume-idol` look and the `古装` register. Shared machinery: `story.md`.
+
+## 1. The promise — winning with the mind
+
+- **智性爽**: watching the heroine read the room, set a trap and flip the verdict. In 琅琊榜 "most
+  strategies rely on predicting how another person will react"; 《冒姓琅琊》 (2025) sold "知识即武器的智性爽感".
+  Revenge rewards a *power* reversal; 宫斗 rewards a *mind* reversal — the payoff is "she saw it
+  coming", not "she turned out to be rich".
+- **升级**: the rank ladder is the progress bar (延禧攻略: becoming stronger alongside her).
+- **道德资本**: the winner must deserve it — "I didn't want to fight; you forced me". Losers lack skill
+  or are exposed doing evil.
+- **A tragic undertow** (optional): 甄嬛 won at the cost of love; 庆余年 is "a 爽文 shell over a cruel
+  truth" — in 短剧, one bittersweet beat near the end.
+
+## 2. The series — schemes, a ladder, factions
+
+| act | ~episodes (of 80) | function |
+| --- | --- | --- |
+| 入局 | 1–10 | low-rank entry (选秀, 庶女, 替身, 重生); the first humiliation and small counter-win; the first rank gain; the antagonist's faction in action; the 卡一 cliff (E8–12) is **a scheme near-exposure** |
+| 立足 | 11–30 | scheme rounds of 3–6 episodes; allies, promotions, a patron; **the antagonist wins at least once** |
+| 低谷 | 30–50 | a big frame-up succeeds (冷宫, demotion, 禁足, a lost child); an ally betrays; she starts the long plant |
+| 反攻 | 50–70 | intel gathered, the antagonist's servants turned; factions realign; she rises faster |
+| 夺权 | 70–end | **the master counter-scheme planted in Acts II–III pays off**; the final verdict, the top rank or the regency |
+
+- **Long plants are the backbone**: 甄嬛传 plants "得一人心" in episode 1 and pays it off dozens later;
+  梅长苏's twelve-year setup unfolds as numbered moves (北燕, the 兰园藏尸案, 南楚); the 滴血验亲 counter-trap
+  begins the empress faction's collapse — **one scheme changes the factions**.
+- **Each scheme across 3–6 episodes**: 埋 (a prop or intel in passing) → 发 (the accusation; a 危机
+  hook) → 危 (the verdict seems certain; a cliff) → 反 (the counter from what was planted; 智商碾压, a 反转
+  hook) → 罚 / 升 (the antagonist punished, her rank changes). **Overlap the arcs**: scheme B is planted
+  during scheme A's reversal.
+- **Two modes of irony**, about half and half: the audience sees the poison go in (will she notice?),
+  or the heroine's plan is withheld (the reveal is the payoff). Too much withholding feels like a cheat.
+
+**Tracked in the outline** (gate `story-palace`): `outline.rankLadder: { track, levels[] low → high }`,
+`outline.rankEvents[]: { char, ep, to }`, `outline.schemes[]: { id, schemer, target, stakes: minor /
+major / master, setupEp, revealEp, outcome: success / countered / backfired, counteredBy: { type:
+intel / prop / ally / counter-trap / authority, ref, plantEp }, flaw, justice? }`,
+`outline.factions[]`, `outline.allegiance[]: { char, ep, from, to, motive, foreshadowEp }`.
+
+| rule | default | why |
+| --- | --- | --- |
+| `scheme-setup` | ≥ 2 / 5 / 20 episodes between setup and reveal (minor / major / master, the master scaled to the run); a master scheme planted in the first half | the long plant is the genre |
+| `no-teleported-evidence` | every counter is planted before the reveal; every countered scheme records the loser's flaw | no evidence out of nowhere; every loss has a cause |
+| `authority-not-rescue` | at most one scheme per phase countered by the emperor alone | the arbiter has an agenda |
+| `competent-antagonist` | an antagonist scheme succeeds against the heroine before the midpoint | a strong enough villain saves a story |
+| `rank-climb` | the first rank event by E12; ≥ N / 15 rises; no 15-episode gap; a low point at 30–60 %; the top rung in the last 10 % | the progress bar and its dip |
+| `faction-realignment` | 2–4 factions; every switch of sides has a motive and foreshadowing ≥ 5 episodes earlier | betrayals are foreshadowed |
+| `moral-capital` | the heroine only schemes against someone who schemed against her first (or records a justice goal) | 道德资本, and regulatory safety |
+
+## 3. The episode
+
+A cliff, a conflict and a reversal every episode (情绪未完、剧情刚翻新、人物状态临界值); the loop is
+冲突 → 行动 → 反转 → 爽点. A trigger or danger episode ends on 危机 / 反转; the reveal episode pays in
+智商碾压 / 打脸 / 复仇 / 身份揭露. Every accusation scene needs a witness chain: the maid → the 太医 → the
+evidence prop.
+
+## 4. Characters
+
+- **The heroine** goes naive → burned → strategist; with 重生 she starts as a strategist and the arc
+  is the *cost* (an ally, a child, love).
+- **The antagonist decides quality** — never a whetstone; every exit should feel regretful.
+- **The emperor** is the prize and the root of all evil (the 欢宜香); an arbiter with his own agenda,
+  never a free rescue. A fresh variant: the emperor as a two-way opponent (《枕边刃》).
+- **Allies play their own games** (安陵容 scoring a point in the chaos); **servants are pieces** — the
+  channel for intel and planted evidence; turning the antagonist's maid is the Act IV move.
+- **Every schemer has a motive** (宫斗 is driven by family, 宅斗 by individual will).
+
+## 5. Beats and freshness
+
+The kit: 麝香 in ointment, the 红麝香珠, 假孕, 滴血验亲, 巫蛊 and poison, the 私通 frame, 选秀, 请安立规矩,
+翻牌子, 冷宫, 禁足, 降位 / 晋封, 落水, 狸猫换太子, forged letters, scents, the 太医 testifying, 嫡庶, 分家, 夺嫡 and
+站队. **Fresh**: knowledge as the weapon (冒姓琅琊); comedy and workplace framing (《我在皇宫搞内卷》, 《谁在说朕
+坏话》); two strong leads with the emperor as the opponent; **reverse a stock trope** (the 滴血验亲 bowl
+rigged by the accuser; the 假孕 exposed by a prop) for a fresh payoff at the same budget.
+
+## 6. Dialogue
+
+A threat phrased as concern ("妹妹身子弱，可要当心"); rank in the pronouns (本宫, 臣妾, 嫔妾, 奴婢);
+punishment that sounds like mercy; truth said in public, meaning in private. Semi-classical but
+readable in a minute; the reversal line is a crisp trap-question ("那这香囊，为何在你宫里的绣娘手里？").
+
+## 7. What fails, and regulation
+
+降智 schemers (找事逻辑); evidence from nowhere; the emperor as deus ex machina; too many consorts too
+early; a rank that never moves; a heroine who only defends; an unforeshadowed betrayal. **Regulation**:
+宫斗 left prime-time satellite TV in 2011 and was targeted again in 2019; NRTA opposes 历史虚无主义; the
+2026 micro-drama measures ban distorting national history and require an AI label per episode;
+critics call 宫斗 anti-intellectual. **In practice**: a fictional dynasty, no real emperors as
+villains, scheming shown as costly, no cruelty to servants for gain, abortion, torture and 房中术 off
+screen, 巫蛊 exposed as a frame-up, never real magic (`story-advice` flags them).
+
+## 8. Sources (read as search summaries)
+
+知乎 on 宫斗剧法则, 女频宫斗 and 滴血验亲; CSDN on 短剧 structure; 界面 / 澎湃 / 36氪 on 宅斗; 豆瓣 后宫指南; 维舟 on
+宫斗 and 道德资本; 腾讯 / 搜狐 on 梅长苏's moves; 新华网 on 宫斗 and personality; 澎湃 / 虎嗅 (王倦); 腾讯 and 光明网
+on 《冒姓琅琊》; 腾讯 红榜 (《我在皇宫搞内卷》); 网易 on 《谁在说朕坏话》; 流媒体网 三点论 and 卡点; 电影界 on 红果
+编剧第一课; 360doc 编剧红宝书; 知乎 黄金手册; 三联 on 《甄嬛传》; 知乎 on 《女相师》; 百度百科 (《枕边刃》); CDT / 虎嗅 on
+15 years of bans; 澎湃 (2019); NRTA 2025-02 notice and 2026-06 draft measures; 澎湃 on 宫斗 and workplace; 腾讯
+/ 搜狐 on 降智; 中新网 (流潋紫); Strange Horizons on 琅琊榜; 界面 on 延禧攻略.