← Files AI Film Pipeline MasterARCHIVED FILE
skills/ai-film-pipeline-master/references/tools/history/coverage-history.md
71.2 KB · Oct 5, 2026 · 18:36 UTC
# Coverage history — archived build notes These paragraphs were the `## Coverage` sections of the craft files when each part was drafted: running totals for a category, the target it was written toward, what remained for the next part, why a target moved. They were true of the draft. They were not true of the file afterwards - a selector or a later part changed every total - and a count check that binds each number to what it counts found them unbound or stale. **Nothing here was edited.** Each block is copied verbatim with the file and line range it came from, so the history is complete and the shipped files carry only numbers that are generated or bound. --- <!-- GENERATED:toc — do not edit by hand. Generated in the source package --> ## Contents - [`phase-01-story-generation/arcs-2.md`](#phase-01-story-generationarcs-2md) - [`phase-01-story-generation/arcs.md`](#phase-01-story-generationarcsmd) - [`phase-01-story-generation/cat-anta-2.md`](#phase-01-story-generationcat-anta-2md) - [`phase-01-story-generation/cat-anta.md`](#phase-01-story-generationcat-antamd) - [`phase-01-story-generation/cat-beat-2.md`](#phase-01-story-generationcat-beat-2md) - [`phase-01-story-generation/cat-beat.md`](#phase-01-story-generationcat-beatmd) - [`phase-01-story-generation/cat-char-2.md`](#phase-01-story-generationcat-char-2md) - [`phase-01-story-generation/cat-char-3.md`](#phase-01-story-generationcat-char-3md) - [`phase-01-story-generation/cat-char-4.md`](#phase-01-story-generationcat-char-4md) - [`phase-01-story-generation/cat-char.md`](#phase-01-story-generationcat-charmd) - [`phase-01-story-generation/cat-end-2.md`](#phase-01-story-generationcat-end-2md) - [`phase-01-story-generation/cat-end.md`](#phase-01-story-generationcat-endmd) - [`phase-01-story-generation/cat-expo-2.md`](#phase-01-story-generationcat-expo-2md) - [`phase-01-story-generation/cat-expo-3.md`](#phase-01-story-generationcat-expo-3md) - [`phase-01-story-generation/cat-expo.md`](#phase-01-story-generationcat-expomd) - [`phase-01-story-generation/cat-info-2.md`](#phase-01-story-generationcat-info-2md) - [`phase-01-story-generation/cat-info-3.md`](#phase-01-story-generationcat-info-3md) - [`phase-01-story-generation/cat-info.md`](#phase-01-story-generationcat-infomd) - [`phase-01-story-generation/cat-open-2.md`](#phase-01-story-generationcat-open-2md) - [`phase-01-story-generation/cat-open.md`](#phase-01-story-generationcat-openmd) - [`phase-01-story-generation/cat-rel-2.md`](#phase-01-story-generationcat-rel-2md) - [`phase-01-story-generation/cat-rel.md`](#phase-01-story-generationcat-relmd) - [`phase-01-story-generation/cat-seq-2.md`](#phase-01-story-generationcat-seq-2md) - [`phase-01-story-generation/cat-seq.md`](#phase-01-story-generationcat-seqmd) - [`phase-01-story-generation/cat-tens-2.md`](#phase-01-story-generationcat-tens-2md) - [`phase-01-story-generation/cat-tens-3.md`](#phase-01-story-generationcat-tens-3md) - [`phase-01-story-generation/cat-tens-4.md`](#phase-01-story-generationcat-tens-4md) - [`phase-01-story-generation/cat-tens.md`](#phase-01-story-generationcat-tensmd) - [`phase-01-story-generation/cat-theme-2.md`](#phase-01-story-generationcat-theme-2md) - [`phase-01-story-generation/cat-theme.md`](#phase-01-story-generationcat-thememd) - [`phase-01-story-generation/cat-time-2.md`](#phase-01-story-generationcat-time-2md) - [`phase-01-story-generation/cat-time.md`](#phase-01-story-generationcat-timemd) - [`phase-01-story-generation/situations.md`](#phase-01-story-generationsituationsmd) - [`phase-01-story-generation/structures.md`](#phase-01-story-generationstructuresmd) - [`phase-02-screenplay-dialogue/cat-comedy-2.md`](#phase-02-screenplay-dialoguecat-comedy-2md) - [`phase-02-screenplay-dialogue/cat-comedy.md`](#phase-02-screenplay-dialoguecat-comedymd) - [`phase-02-screenplay-dialogue/cat-dial-2.md`](#phase-02-screenplay-dialoguecat-dial-2md) - [`phase-02-screenplay-dialogue/cat-dial.md`](#phase-02-screenplay-dialoguecat-dialmd) - [`phase-02-screenplay-dialogue/cat-emo-2.md`](#phase-02-screenplay-dialoguecat-emo-2md) - [`phase-02-screenplay-dialogue/cat-emo-3.md`](#phase-02-screenplay-dialoguecat-emo-3md) - [`phase-02-screenplay-dialogue/cat-emo.md`](#phase-02-screenplay-dialoguecat-emomd) - [`phase-02-screenplay-dialogue/cat-pov-2.md`](#phase-02-screenplay-dialoguecat-pov-2md) - [`phase-02-screenplay-dialogue/cat-pov.md`](#phase-02-screenplay-dialoguecat-povmd) - [`phase-02-screenplay-dialogue/cat-rhythm-2.md`](#phase-02-screenplay-dialoguecat-rhythm-2md) - [`phase-02-screenplay-dialogue/cat-rhythm-3.md`](#phase-02-screenplay-dialoguecat-rhythm-3md) - [`phase-02-screenplay-dialogue/cat-rhythm.md`](#phase-02-screenplay-dialoguecat-rhythmmd) - [`phase-02-screenplay-dialogue/cat-scene-2.md`](#phase-02-screenplay-dialoguecat-scene-2md) - [`phase-02-screenplay-dialogue/cat-scene-3.md`](#phase-02-screenplay-dialoguecat-scene-3md) - [`phase-02-screenplay-dialogue/cat-scene.md`](#phase-02-screenplay-dialoguecat-scenemd) - [`phase-02-screenplay-dialogue/cat-sub-2.md`](#phase-02-screenplay-dialoguecat-sub-2md) - [`phase-02-screenplay-dialogue/cat-sub.md`](#phase-02-screenplay-dialoguecat-submd) - [`phase-02-screenplay-dialogue/cat-voice-2.md`](#phase-02-screenplay-dialoguecat-voice-2md) - [`phase-02-screenplay-dialogue/cat-voice.md`](#phase-02-screenplay-dialoguecat-voicemd) - [`phase-05-song-generation/cat-lyric-2.md`](#phase-05-song-generationcat-lyric-2md) - [`phase-05-song-generation/cat-lyric-3.md`](#phase-05-song-generationcat-lyric-3md) - [`phase-05-song-generation/cat-lyric.md`](#phase-05-song-generationcat-lyricmd) - [`phase-01-story-generation/formats-social.md`](#phase-01-story-generationformats-socialmd) - [`phase-01-story-generation/intents.md`](#phase-01-story-generationintentsmd) - [`phase-01-story-generation/symptoms.md`](#phase-01-story-generationsymptomsmd) - [`phase-02-screenplay-dialogue/scenes-2.md`](#phase-02-screenplay-dialoguescenes-2md) - [`phase-02-screenplay-dialogue/scenes.md`](#phase-02-screenplay-dialoguescenesmd) <!-- /GENERATED:toc --> ## `phase-01-story-generation/arcs-2.md` *Lines 341–341:* Running total for character arcs: **52** — 40 in `arcs.md`, 12 here. *Lines 353–353:* **Still missing, recorded rather than hidden.** Three, and none of them is small. **The arc in translation and across cultures** — whether a change arc reads as growth or as betrayal depends on whether the audience's culture prizes individuation or obligation, and nothing in this database says so; `char.inside_outside_culture` and `theme.untranslatable_value` are adjacent and are not it. **The arc of the audience in non-interactive work** — the designed change in a reader or viewer that advocacy, satire and PSA are actually built to produce; `arc.player_arc` covers only the interactive case, and `intents.md` routes the effect without a card for the curve. **Interactive non-fiction arcs** — branching documentary and chosen testimony, where a real person's change is steered by a viewer, which is also the largest open hole in [`interactive-and-branching.md`](../../phase-01-story-generation/interactive-and-branching.md)'s coverage — `inter.player_driven_arc` and `inter.visitor_arc` are adjacent and neither is the documentary case. Each is worth two to four cards, which would take character arcs to roughly 60. --- ## `phase-01-story-generation/arcs.md` *Lines 730–730:* **Running the arc: the mechanics of change** (7) — where the four turns sit against structure, the first serious challenge to the Lie, the win that rewards the Lie, the point of no return, the self-revelation, the choice that *is* the change, and the proof scene. The self-revelation and the choice are deliberately two cards, following Truby: knowing and doing are different beats and most weak endings collapse them. *Lines 732–732:* **Arcs that are not one of the five** (5) — the growth arc with no Lie (which absorbs the education arc, since they run on the same machinery), the acceptance arc, the decision arc, the double arc and the group arc. *Lines 734–734:* **Arcs by unit** (5) — episode against season, the borrowed arc that lets a procedural lead stay fixed, the trilogy and long-series arc, the arc under sixty seconds, and the arc in a lyric. *Lines 736–736:* **Ensemble and secondary arcs** (2) — the supporting character's arc and its three-scene budget, and arc collision. *Lines 738–738:* **Working method and diagnosis** (3) — writing the arc backwards from the final choice, the arc statement as a one-line test, and the four-scene arc audit. *Lines 740–740:* **Named failures** (5) — `arc.told_revelation`, `arc.plot_solved_arc`, `arc.unattacked_lie`, `arc.midpoint_peak`, `arc.padding_relapse`. All carry `formats: — | genres: —`. --- ## `phase-01-story-generation/cat-anta-2.md` *Lines 646–646:* **30 cards** in this file, across 9 sections: The antagonist across a series (5), The monster and the horror antagonist (4), Comedy and rivalry (3), Kids and the bounded opponent (4), Opposition the audience controls (2), Opposition drawn from life (3), Sympathy and redemption (4), Resources and tiers (3), Opposition Without a Person (3). *Lines 648–648:* Running total for category `anta`: **57 of 57** — 27 in `cat-anta.md`, 30 here. Every item named in Part 1's Coverage list is now written. *Lines 650–650:* The target moved from 55 to 57. Three techniques surfaced during the research that the original estimate had no room for — `anta.fate`, `anta.the_crowd` and `anta.no_author` — and on the README's own terms the target moves rather than the craft being cut to fit it. One card was also removed after it was written: What the Monster Is About duplicated `theme.monster_meaning`, so its unique content was folded into that card and only a cross-reference is kept here. One home per technique. --- ## `phase-01-story-generation/cat-anta.md` *Lines 434–434:* **Still to write in this category:** the antagonist across a series and how to retire one, the recurring versus episodic opponent, opposition in procedural formats, the monster and its rules, the antagonist in horror specifically, opposition in comedy, the antagonist in kids' content, opposition in interactive work where the player chooses, the sympathetic villain and its limits, opposition drawn from real people and institutions, the redeemed antagonist, and the antagonist's own arc. Target for the full category: 55. --- ## `phase-01-story-generation/cat-beat-2.md` *Lines 930–930:* **38 cards** in this file, across 10 sections: Ensemble and Multi-Handed Beats (4), Mediated, Off-Screen and Absent Beats (5), Narration, Verse and Song (4), Beats in Other Media Grammars (5), Handoff, Pressure and Real Time (3), The Solo Beat and the Environment (3), Comic Beat Forms Beyond the Third (3), The Decision and the Exposition Scene (2), Testing and Notating Beats (3), Further Named Failures (6). *Lines 932–932:* Running total for category `beat`: **70 of 70** — 32 in `cat-beat.md`, 38 here. *Lines 934–934:* Everything Part 1's Coverage section listed for Part 2 is present, plus four techniques the research added: the beat carried by the cut, the crowd reaction beat, the task beat, and the two-second beat of vertical short form. One item considered and deliberately excluded: the animatic/storyboard beat, which on inspection is a production timing unit rather than a writing unit and is already served by `beat.panel_gutter` for static sequencing and by the `seq` category for run-time planning. The category is closed at 70; nothing further is outstanding. --- ## `phase-01-story-generation/cat-beat.md` *Lines 697–697:* 32 cards. This is Part 1 of 2 toward a 70-card target for category `beat`; 38 cards remain. --- ## `phase-01-story-generation/cat-char-2.md` *Lines 695–695:* **40 cards** in this file, across nine sections: Whose Story It Is (5), Belief and Worldview (5), Work, Class and Money (5), The Character Under Pressure (5), Change and Its Absence (4), Short and Serial Forms (4), Culture, Faith and Place (4), The Ensemble's Shape (4), Named Failures (4). *Lines 697–712:* Running total for category `char`: **74 of 120** — 34 in `cat-char.md`, 40 here. **46 remain** for Parts 3 and 4. Still to be written: - **Voice and idiolect as character** — the construction of a speech signature: vocabulary range, sentence length, what they never say, profanity budget, silence as a trait. - **Bodies and constraint** — illness, disability, pregnancy, exhaustion, addiction, hunger, injury, written as conditions of action rather than as pity levers. - **Sexuality and desire for another person** as distinct from the story want: jealousy, courtship behaviour, seduction competence. - **Character in groups** — the family system, birth order, the workplace hierarchy, gangs and units, the crowd member. - **Non-human and near-human characters** — animals, machines, AI, gods, the personified place, the choral narrator, children's-animation species logic. - **Duplicates and doubles** — twins, impostors, disguises, the undercover identity, the character playing another character. - **Names in translation and heritage naming** — transliteration, honorifics, kinship terms of address. - **Character bibles and series maintenance** — the show-bible entry, continuity of behaviour across a writers' room, recasting, the character rewritten for a new actor. - **Casting-adjacent craft** — writing for a known actor, writing against an actor's image, age ranges, how a role reads to an agent. - **Development tools** — the character interview, the wound questionnaire and its limits, the two-column want/need sheet, the audition scene. - **Comedy character construction** — the comic flaw, the straight man, escalation of a trait, the character who is always wrong. - **Deaths, exits and absences beyond design** — the off-screen death, the survivor, the character who leaves alive, the ghost still present in every scene. - **The class tell and the daily-labour beat** — inherited versus earned ease, the working day as structure. - **Children and animals as scene partners**, and the character written to be played by a non-actor. --- ## `phase-01-story-generation/cat-char-3.md` *Lines 876–876:* **44 cards** in this file, across nine sections: Voice and Idiolect as Character (4), Bodies, Conditions and Constraint (8), Desire for Another Person (4), The Character Inside a Group (4), Non-Human and Near-Human Characters (5), Duplicates, Doubles and Assumed Identities (4), Names, Translation and Heritage Naming (5), Bibles, Series Maintenance and Casting (5), Development Tools (5). *Lines 878–878:* Running total for category `char`: **118** — 34 + 40 + 44. The estimate was 120, and the brief for this file set 38 and a close at 112. *Lines 880–880:* **The target moved, and here is the honest reason.** The ten outstanding areas were enumerated card by card before drafting, against the existing 74 and against `arcs.md`, `cat-voice*.md`, `cat-dial*.md`, `cat-rel*.md` and `cat-sub*.md`. After removing everything that already had a home, 44 distinct techniques remained — and that number is already the product of three deliberate merges, not an expansion: the impostor, the long-term undercover identity and the worn disguise are one card (`char.assumed_identity`) rather than three, because the maintenance craft is identical; the robot and the artificial mind are one card (`char.machine_character`); and the crowd member absorbed what would otherwise have been a separate card on the individual inside an institution. Cutting to 38 would have meant dropping six real techniques, which the README forbids. Two remaining gaps are named below rather than padded into existence. *Lines 900–900:* **Genuinely still absent from the category, named rather than padded:** two items from Part 2's list did not survive the one-home rule and have no card anywhere. The first is **the character written to be played by a non-actor** — real people, children and first-time performers in heritage and documentary drama, where the writing constraint is that the part must be sayable by someone who cannot act, which is a real technique with nothing covering it. The second is **the class tell as inherited versus earned ease** — `char.cannot_afford` and `char.money_specifics` cover poverty's arithmetic, and nothing covers the opposite, the unexamined ease of someone who has never had to ask a price. Both are one card each; the category could reasonably sit at 120 with them in it, which is where the original estimate landed and, in hindsight, for roughly the right reasons. --- ## `phase-01-story-generation/cat-char-4.md` *Lines 200–200:* **10 cards** in this file. Running total for category `char`: **128** — 34 + 40 + 44 + 10. *Lines 208–208:* **The two gaps `cat-char-3.md` named are now closed.** `char.written_for_a_non_actor` and the inherited-ease pair (`char.inherited_ease`, `char.class_tell_survives_money`) were listed in that file's Coverage as "genuinely still absent from the category, named rather than padded", with the estimate that the category could reasonably sit at 120 with them in it. It sits at 128, because the non-actor card needed its failure twin and the class material needed two cards rather than one — inherited-versus-earned ease and the residue that survives a change of circumstances are different instruments with different tests, and merging them would have lost the asymmetry rule, which is the useful part. --- ## `phase-01-story-generation/cat-char.md` *Lines 576–576:* **34 cards** in this file, across eight sections: Entrances and First Impressions (6), Desire / Wound / Lie (5), Contradiction and Dimension (4), Body / Habit / Object / Gaze (5), Backstory, Secret and Name (3), Relationships, Status and Ensemble (5), Antagonist Construction (2), Ages, Real People and Traps (4). *Lines 578–596:* Category target is **120 cards**, so **86 remain** for parts 2 to 4. Still to be written: - **Voice and idiolect as character** (not dialogue technique — the construction of a speech signature: vocabulary range, sentence length, what they never say, code-switching, profanity budget, silence as a trait). - **Belief systems**: faith, ideology, superstition, professional code, and how to write a believer without endorsing or mocking. - **Bodies and constraint**: illness, disability, pregnancy, exhaustion, addiction, hunger, injury — written as conditions of action rather than as pity levers. - **Class, money and work**: occupation as characterisation, what a character earns and spends, the daily-labour beat, the hands test. - **Sexuality and desire for another person** as distinct from the story want; jealousy, courtship behaviour, the seduction competence. - **Character in groups**: the family system, birth order, the workplace hierarchy, gangs and units, the crowd member. - **Non-human and near-human characters**: animals, machines, AI, gods, the personified place, the choral narrator, children's-animation species logic. - **Duplicates and doubles**: twins, impostors, disguises, the undercover identity, the character who is playing another character. - **Names in translation and heritage naming** across languages, transliteration, honorifics, kinship terms of address. - **Character bibles and series maintenance**: the show bible entry, continuity of behaviour across a writers' room, recasting, the long-running character who must not change. - **Casting-adjacent craft**: writing for a known actor, writing against an actor's image, age ranges, how a role reads to an agent. - **Development tools**: the character interview, the wound questionnaire and its limits, the two-column want/need sheet, the audition scene. - **Ensemble economics**: how many characters a runtime can hold, doubling and combining characters in the rewrite, the character who should be cut. - **Deaths, exits and absences**: writing the character who leaves, the ghost who is still in every scene, the off-screen death, the survivor. - **Culture-specific construction**: writing outside your own community, research obligations, the consultant, the difference between specificity and stereotype. - **Comedy character construction**: the comic flaw, the straight man, the character who is always wrong, escalation of a trait, the sitcom character who cannot learn. - **Short-form character**: building a person in 15 seconds, the vertical-drama archetype set, the recurring-creator persona, character in a hook. --- ## `phase-01-story-generation/cat-end-2.md` *Lines 739–739:* 39 cards in Part 2. Running total for category `end`: **63 of 70** (24 in `cat-end.md`, 39 here). *Lines 743–743:* **Still to write toward 70 (7 remaining):** the ending in adaptation when the audience already knows it; endings under censorship or a commissioner's note; the anthology ending and how a collection closes; the stage ending and the curtain; the ending of a work with two protagonists; the comic-book and comics-page ending; and the ending that has to work in two languages. --- ## `phase-01-story-generation/cat-end.md` *Lines 395–395:* **Still to write in this category:** genre-specific ending conventions (the romance ending, the crime ending, the horror coda, the sports final), the tragic ending and its rules, endings in interactive and branching work, multiple endings by design, series finales as distinct from season finales, the ending in kids' content, credits and post-credits structure, and endings for non-narrative forms. Target for the full category: 70. --- ## `phase-01-story-generation/cat-expo-2.md` *Lines 485–485:* Running total for category `expo`: **60 of 60** — 32 in `cat-expo.md`, 28 here. --- ## `phase-01-story-generation/cat-expo-3.md` *Lines 334–334:* **12 cards** in this file. Running total for category `expo`: **72** — 32 + 28 + 12. *Lines 338–338:* **Item 10 is already written, three times, and better.** The brief asked for the card that the season-two problem generalises into — a reader or viewer returning after months. The database already holds the whole of it for narrative: `info.between_episode_memory` carries the decay model, and it is a real one (emotional states survive a gap, names decay, plot mechanics collapse, and binge audiences invert the middle two); `expo.season_two_reteach` carries the delivery, including the rule that you re-teach only the rules the new season will bend; `expo.re_entry_exposition` carries the weekly form. A fourth narrative card would have been a duplicate with a different name. What is genuinely absent is the *non-narrative* case — the reader who did half a procedure six months ago — and that is written here as `expo.lapsed_learner_re_entry`, which points at all three rather than competing with them. The useful finding is that the sorting rule inverts between the two: a story audience loses facts and keeps feeling; an instruction reader loses state and keeps the concept. *Lines 340–340:* **Item 9 is exposition, not rhythm, and the answer turned out to be sharper than the question.** Writing to a decodable or levelled specification is not primarily a constraint on sentence music — it is a constraint on *subject*. If the phonics sequence you are writing to has not reached long vowels, you cannot write about the sea, and the discovery that your subject is unwriteable arrives at draft three. That is a delivery problem and it belongs here, as `expo.decodability_spec`. The rhythm half of it — the sentence-length ceiling and what the mandated repetition does to the ear — is genuinely a rhythm matter and is flagged here for whoever holds that file; the two cards would not overlap. --- ## `phase-01-story-generation/cat-expo.md` *Lines 618–618:* **32 cards** in this file: 4 on the information budget, 5 on delivery by demonstration, 5 on terminology and apparatus, 4 on visual and environmental delivery, 3 on placement and openings, 5 on nonfiction, technical and short-form, 3 named failures, 3 repair techniques. *Lines 620–628:* Toward a 60-card target for `expo`, **28 remain** for Part 2. Not yet covered here: - Genre-specific delivery: horror rule-setting, comedy exposition (the fast joke that also informs), children's and animation exposition, sports and game-rule delivery, adaptation exposition where the audience is assumed to know the source. - Channel techniques: voice-over strategies and their contracts, the interview cutaway, on-screen graphics and kinetic text, title cards and crawls, chapter headings and epigraphs, found documents and epistolary delivery. - Reinforcement and redundancy: the two-touch rule for audio-first formats, deliberate re-delivery through a second channel, the read-aloud test. - Series-scale problems: pilot-load versus season-load, re-entry exposition at episode openings, the recap sequence, staggered audience knowledge across an ensemble, re-teaching rules to season-two viewers. - Interactive and game-specific: tutorial design, the codex entry, ambient NPC exposition, exposition the player can skip. - Further failures: the exposition-fairy character, the flashback that explains what was already clear, the over-taught rule, retroactive rule invention, the recap that spoils, translation-convention breaches. - Further repairs: converting a statement into a question the character asks, splitting one block across three scenes, trading a paragraph of exposition for a single image. --- ## `phase-01-story-generation/cat-info-2.md` *Lines 748–748:* **This file contains exactly 37 cards.** Running total for category `info`: 32 (Part 1) + 37 (Part 2) = **69 of the target 90. 21 cards remain for Part 3.** *Lines 752–760:* **Remaining for Part 3 — 21 cards.** Take them from Part 1's outstanding list, minus what Part 2 has now absorbed: - **Reveal variants still open:** the confession, the overheard reveal, the third-party reveal, the reveal by absence, the reveal the audience already guessed and how to pay it anyway, the deliberately anticlimactic reveal, the refused reveal, the delayed denial. - **Withholding mechanics:** the ellipsis and the cut-away, the offstage catastrophe, the redacted document, the untranslated line, the obscured face, the withheld name, the withheld year, the deliberately ambiguous image. - **Structural information devices:** the framing device and its narrator, the nested account, the found-footage constraint, the epistolary gap, multiple-POV rotation and its handover rules, flashback as evidence, the flash-forward contract, the parallel-timeline reveal, the time-loop information rule. - **Genre-specific control:** the horror rule of the unseen, comic dramatic irony and farce mechanics, the romance misunderstanding and why most of them cheat, courtroom disclosure, the heist withhold-the-plan convention and its abuse, war-film information fog. - **Audience-position management:** deliberate position switching mid-story and its cost, the ensemble information map, sustaining a gap across a long serial, re-opening a closed gap, running two positions at once. - **Failure diagnostics still open:** the dead plant, the orphan payoff, the exposition dump disguised as a reveal, the twist that only works once, the reveal that arrives after the audience stopped caring. - **Short-form and interactive:** the loop-bait reveal, the comment-section reveal, the reveal designed for rewatch, information control under autoplay, player-discovered information, the optional-lore problem, reveal design for non-linear consumption. --- ## `phase-01-story-generation/cat-info-3.md` *Lines 285–285:* **10 cards** in this file. Running total for category `info`: **79** — 32 + 37 + 10. --- ## `phase-01-story-generation/cat-info.md` *Lines 557–565:* **Remaining for category `info` — target 90, so 58 cards outstanding across Parts 2 and 3:** - **Reveal variants:** the confession, the document reveal, the overheard reveal, the third-party reveal, the reveal by absence, the reveal the audience already guessed (and how to pay it anyway), the deliberately anticlimactic reveal, the refused reveal, the reveal that never comes, the delayed denial. - **Withholding mechanics:** the ellipsis and the cut-away, the unseen event, the offstage catastrophe, the redacted document, the untranslated line, the obscured face, the withheld name, the withheld year, the unreliable timeline, the deliberately ambiguous image. - **Structural information devices:** the framing device and its narrator, the nested account, the contradictory-testimony structure, the found-footage constraint, the epistolary gap, multiple-POV rotation and its handover rules, flashback as evidence, the flash-forward contract, the parallel-timeline reveal, the time-loop information rule. - **Genre-specific control:** the horror rule of the unseen, comic dramatic irony and farce mechanics, the romance misunderstanding and why most of them cheat, courtroom disclosure, the heist withhold-the-plan convention and its abuse, war-film information fog, the historical audience-already-knows problem, the documentary reveal and archive ethics. - **Audience-position management:** deliberate position switching mid-story, the cost of each switch, the ensemble information map, sustaining a gap across a long serial, re-opening a closed gap, running two positions simultaneously. - **Failure diagnostics:** the dead plant, the orphan payoff, the double-told fact, the reveal with no receiver, the exposition dump disguised as a reveal, the twist that only works once, the mystery nobody asked about, the reveal that arrives after the audience stopped caring. - **Short-form and interactive:** the loop-bait reveal, the comment-section reveal, the reveal designed for rewatch, information control under autoplay, player-discovered information in game narrative, the optional-lore problem, reveal design for non-linear consumption. --- ## `phase-01-story-generation/cat-open-2.md` *Lines 723–723:* 38 cards in Part 2. Running total for category `open`: **66 of a target 70**. *Lines 727–727:* **Still to write in Part 3:** the season opening and the return after a hiatus, the trailer as an opening in its own right, the opening of an adaptation for a double audience, the stage play's first minutes before the audience has settled, the podcast episode open, the re-open after an advertising break, and a diagnostic pass for auditing an existing opening against its ending. That is roughly eight cards, which will bring the category slightly past the 70 target rather than exactly to it. --- ## `phase-01-story-generation/cat-open.md` *Lines 457–457:* **Still to write in this category:** openings by genre convention (the horror cold open, the romance meet, the heist assembly, the procedural body), the prologue and its legitimate uses, opening lines as a craft unit, the first paragraph in prose, title and logline construction, cover and thumbnail as an opening, the first page of a pitch document, and the opening in translated or subtitled work. Target for the full category: 70. --- ## `phase-01-story-generation/cat-rel-2.md` *Lines 539–539:* **28 cards** in this file, across six sections: Distance, Absence and the Barrier (7), Care, Obligation and the Third Party (6), Beginnings, Ruptures and Repairs (5), Three or More (5), The Ensemble on the Page (2), Named Failures (3). *Lines 543–543:* Running total for category `rel`: **60 of 60** — 32 in `cat-rel.md`, 28 here. *Lines 545–545:* Two genuine areas remain unwritten and are recorded rather than hidden, so a later editor can decide whether the target moves: **the relationship with an institution's own members across a power gap that cannot be closed** (the soldier and the officer, the parishioner and the priest — currently absorbed into `rel.institutional_bond` and `rel.group_hierarchy_display`, arguably its own card), and **the ensemble conducted asynchronously**, where the group never assembles at all and exists only as a thread, a channel or a rota. If either is written, the category target moves to 62; nothing here was padded to reach 60. --- ## `phase-01-story-generation/cat-rel.md` *Lines 654–654:* Toward a 60-card target, Part 2 should carry: the reunion after long separation; the relationship conducted entirely by letter, text or phone; the bond with a dead person; the relationship across a language barrier; the pair separated by class; the caretaker and the cared-for; the pair with a shared child; first meetings and how to write a non-meet-cute; the architecture of the breakup scene; the reconciliation that does not restore the old terms; the third wheel; the group that expels a member; the succession fight; the ensemble that loses its leader; the non-human member; the institutional relationship (a person and their employer, church or army); parasocial and one-sided bonds; the pair who are wrong about what their relationship is; jealousy as a three-person structure; relationships across a time jump; the ensemble pilot and how to introduce eight people at once; the reunion ensemble; the relationship that only works in one location; and three further failures — the group that exists only to admire the protagonist, the rupture with no aftermath, and the reconciliation that arrives one scene after the rupture. --- ## `phase-01-story-generation/cat-seq-2.md` *Lines 637–637:* Running total for category `seq`: **60 of 60** — 32 in `cat-seq.md`, 28 here. --- ## `phase-01-story-generation/cat-seq.md` *Lines 747–753:* **32 cards in this file**, across five sections: - Section 1 — What a Sequence Is: 7 cards - Section 2 — Building and Joining: 6 cards - Section 3 — Sequence Types: 8 cards - Section 4 — Rhythm and Placement: 6 cards - Section 5 — Diagnosis and Repair: 5 cards (4 named failures) *Lines 755–755:* Toward the 60-card target for category `seq`, Part 2 should cover: the sequence outline as a working document and how to write one before drafting; sequence-level reversals and the mid-run turn; the sequence in relation to subplot (how a B-story sequence interleaves with an A-story sequence); sequence openings and how much context a run needs before it can start; the cold-open sequence; the flashback sequence and the embedded-past run; the dream or imagined sequence and its cost in trust; the trial or hearing sequence; the negotiation sequence; the siege and the containment run; the escape sequence as the inverse of the infiltration sequence; the ritual sequence (wedding, funeral, ceremony) as a structured container; comic sequences and the escalation of a running gag across scenes; the horror stalk sequence and its beat pattern; sequences in animation and kids' content, where run length is constrained by attention; sequences in games, where the player controls duration; sequences in documentary and non-fiction narrative; sequence-level foreshadowing and payoff bookkeeping; how sequences carry theme without stating it; the relationship between sequence count and running time across formats; adapting sequences between media; and the remaining failure modes — the repeated sequence (two runs asking the same question), the orphaned sequence (a run whose outcome never returns), the premature climax, the sequence that resolves fully and stops the story, and the false convergence. --- ## `phase-01-story-generation/cat-tens-2.md` *Lines 728–728:* **Cards in this file: 38.** Running total for category `tens`: **69** (31 in Part 1 + 38 here). Target is 80, so **11 remain for Part 3.** *Lines 732–740:* Remaining for Part 3 (11 cards), drawn from the clusters Part 1 opened: - **Hiding and breath-holding** — the concealment beat, the crowd as cover, the pursuit that pauses. - **Threat visibility** — the monster-reveal clock, the threat shown too early, the second antagonist. - **Surveillance and the watched life** — being observed as an ongoing state rather than an event. - **Wrong places and ritual** — the site that is wrong, ritual tension, dream logic and what it costs. - **Post-climax and ensemble** — the aftershock beat, and tension in multi-strand and ensemble stories. - **Remaining medium mechanics** — tension in documentary, in animation and family work, and in subtitled or dubbed delivery. - **Remaining repairs** — over-telegraphing, the unearned reprieve, and rebuilding a second act that has no clock. --- ## `phase-01-story-generation/cat-tens-3.md` *Lines 199–199:* **10 cards** in this file. Running total for category `tens`: **79** — 31 + 38 + 10. *Lines 213–213:* **Honestly outstanding.** Part 2 set the category target at 80 and listed seven clusters for a Part 3: hiding and breath-holding, threat visibility, surveillance as a state, wrong places and ritual, post-climax and ensemble, remaining medium mechanics, and the remaining repairs. **None of those are in this file** — this Part 3 was written to close the router's reported gaps instead, and those clusters are still unwritten. They are real cards and would make a Part 4 of roughly twenty. The running total is 79 against a target of 80, but the one-card shortfall is arithmetic rather than meaningful: the target should move when that Part 4 is written. --- ## `phase-01-story-generation/cat-tens-4.md` *Lines 394–394:* Running total for category `tens`: **99** — 31 + 38 + 10 + 20. The estimate was 80, and the target moves. It was set from an architecture sketch before a single card existed, and the seven clusters `cat-tens-2.md` reserved turned out to hold nineteen or twenty real techniques rather than the eleven the arithmetic had left room for. Nothing here was written to reach a number and nothing real was cut to stay under one. What is genuinely still absent is named below. *Lines 405–405:* **Honestly outstanding.** Three items from the reserved clusters are not here and are not cited elsewhere. **The pursuit that pauses** — the hunter who stops moving and converts a chase into a siege — is a real card and was cut for space; `tens.waiting_as_weapon` and `seq.stalk_sequence` cover its edges but neither owns the moment the chase changes state. **Dream logic and what it costs** is likewise real, likewise unwritten, and belongs with `tens.almost_right_place`. **Tension in documentary, in animation and family work, and in subtitled delivery** were listed by Part 2 as medium mechanics; the brief for this part replaced them with prose, audio, comics and stage, and the replacement was the right call, but the three remain unowned by `tens` — `seq.documentary_sequence`, `dial.subtitle_constraint` and `rhythm.subtitle_beat` cover their own axes and none of them covers fear. If a Part 5 is ever written it is those five cards and no more, and the category would close at 104. --- ## `phase-01-story-generation/cat-tens.md` *Lines 522–532:* Target for category `tens` is 80 cards. Remaining for Parts 2 and 3 (49 cards): - **Chase, pursuit and physical suspense** — the slow pursuer, the near-miss, the hiding beat, the pursuit that pauses, breath-holding scenes, the crowd as cover, vehicle suspense. - **Threat design** — the off-screen antagonist, the threat that is never shown, the threat that is shown too early, scale mismatch, the threat with a face vs the threat with a rule, contagion logic, the second antagonist. - **Paranoia and trust tension** — the traitor in the group, the test of loyalty, the character who might be imagining it, gaslighting pressure, the witness nobody believes, surveillance dread. - **Body, gore and restraint** — implication over depiction, the sound of the injury, the aftermath shot, medical dread, the wound that gets worse, disgust vs fear. - **Supernatural and cosmic dread** — the indifferent force, scale horror, the thing that cannot be named, ritual tension, the place that is wrong, dream logic and its costs. - **Comedy and tension** — suspense in comedy, the escalating lie, farce pressure, the awkward pause as engine, comic dread, tension in children's and family work. - **Tension across the whole structure** — the tension map, act-level pressure curves, the false victory, the second-act dip, pre-climax stillness, the post-climax aftershock, tension in ensemble and multi-strand stories. - **Delivery and craft mechanics** — sentence length as pulse, paragraph breaks as heartbeat, tense and person choices, POV distance and fear, tension in voiceover, tension in subtitled and dubbed work, tension in animation, tension in documentary, tension in stage blackouts, tension in game loops and player agency. - **Failure modes and repair** — diagnosing a flat scene, tension fatigue, over-telegraphing, the unearned reprieve, the audience that stopped believing the threat, repairing a second act with no clock. --- ## `phase-01-story-generation/cat-theme-2.md` *Lines 649–649:* **30 cards** in this file, across 10 sections: Comedy, satire and irony (4), Allegory and fable (3), Theme for children and young audiences (3), Brand, advertising and commissioned work (3), Non-fiction and interactive work (2), Genre and form carrying the claim (2), Adaptation and translation (4), Culture and faith, inside and outside (4), Theme and the character arcs (2), Ambiguity and the shape of the verdict (3). *Lines 651–651:* One card carries a review date: `theme.purpose_washing`. The other 29 are durable. *Lines 653–653:* Running total for category `theme`: **55 of 55** — 25 in `cat-theme.md`, 30 here. Every item named in Part 1's Coverage section is now written. *Lines 655–655:* Three areas remain genuinely thin and would justify moving the target rather than padding it: **theme in interactive work** has one card here where the craft supports three or four (the argument made by a mechanic rather than a choice, the ending set as a spread of positions on one question, and theme in a conversational agent, which has no film equivalent at all); **theme in short and vertical form**, where the claim has to land inside thirty seconds and the usual instruments — motif, ensemble, arc — are all unavailable; and **theme in stand-up and monologue**, where the argument is conducted by a persona in the first person and the anti-theme has to be voiced by the same mouth. If a Part 3 is written, those are its contents, and the category target moves to roughly 65. --- ## `phase-01-story-generation/cat-theme.md` *Lines 411–411:* **Still to write in this category:** theme in comedy and satire, theme in kids' and preschool content, theme in advertising and brand work, allegory and fable construction, irony as a thematic instrument, ambiguity and how much to leave open, cultural and religious material handled from inside versus outside, theme in adaptation, theme across translation, and the relationship between theme and the five character arcs. Target for the full category: 55. --- ## `phase-01-story-generation/cat-time-2.md` *Lines 571–571:* **33 cards** in this file, across seven sections: Tense as a Time Device (3), Whole-Story Time Shapes (7), Dated Records and Contested Time (4), Memory Inside Memory (3), Pause, Pace and the Reader's Clock (5), Tracking Time Across a Long Work (5), and Named Failures (6). *Lines 575–575:* Running total for category `time`: **65 of 65** — 32 in `cat-time.md`, 33 here. Part 1's header says "Part 1 of 3"; the category in fact closes in two files, and this is the second. There is no Part 3, and nothing from Part 1's reserved list remains outstanding. --- ## `phase-01-story-generation/cat-time.md` *Lines 519–519:* **32 cards** in this file: 8 on arrangement (linear, in medias res, non-linear assembly, reverse chronology, frame narrative, dual timeline, parallel timelines, anniversary structure), 8 on the flashback and flashforward, 6 on duration (scene, summary, ellipsis, dilation, real time, iterative), 6 on signals, montage and format-specific chronology, and 4 named failures. *Lines 521–521:* Toward a 65-card target for this category, **33 remain**, and they are all written — `cat-time-2.md` closes the category in two files, not three. Reserved for Part 2 and now delivered there: tense choice as a time device (present vs past vs future, retrospective distance, the shifting-tense novel); narrative pause and descriptive stopping; the time loop and its rules; time travel chronology and paradox management; prophecy and foreshadowed futures as temporal structure; the countdown chapter; multi-generational and century-spanning structure; epistolary and document-dated chronology; diary and log-file time; the single-day novel; the seasonal cycle; simultaneity and the "meanwhile" problem in prose; retroactive continuity; chronology in translation and across calendars; flashback within flashback; the memory that belongs to someone else; deathbed and life-review structures; pace-control by chapter length; the temporal cliffhanger; time in oral and performed storytelling; ageing and de-ageing a character on the page; chronology of an ensemble; the false flashforward; the dream sequence distinguished from the flashback; historical present used for the past; the timeline audit as a revision tool; time in documentary and non-fiction narrative; and the reader-facing timeline aid (family tree, date header, map) and when it is an admission of failure. --- ## `phase-01-story-generation/situations.md` *Lines 727–733:* | Set | Cards | |---|---:| | Polti's 36 dramatic situations | 36 | | Propp's 31 functions plus the 7 roles | 32 | | Booker's 7 basic plots | 7 | | Tobias's 20 master plots | 20 | | **Total** | **95** | --- ## `phase-01-story-generation/structures.md` *Lines 715–728:* | System | Beats | ids | |---|---:|---:| | Three-Act Paradigm | 7 | 8 | | Save the Cat | 15 | 16 | | Story Circle | 8 | 9 | | Hero's Journey | 12 | 13 | | 22 Building Blocks | 22 | 23 | | Sequence Approach | 8 | 9 | | Freytag's Pyramid | 5 | 6 | | Kishōtenketsu | 4 | 5 | | Story Spine | 6 | 7 | | Four-Act | 5 | 6 | | McKee's Hierarchy | 6 | 7 | | **Total** | **98** | **109** | --- ## `phase-02-screenplay-dialogue/cat-comedy-2.md` *Lines 921–921:* **38 cards in this file.** With Part 1's 32, the running total for category `comedy` is **70 of 85**, so **15 remain** for Part 3. --- ## `phase-02-screenplay-dialogue/cat-comedy.md` *Lines 672–672:* **32 cards in this file.** The target for category `comedy` is 85, so **53 remain** across Parts 2 and 3. --- ## `phase-02-screenplay-dialogue/cat-dial-2.md` *Lines 787–787:* **39 cards** in this file. Running total for category `dial`: **70 of 130** (31 in Part 1, 39 here). *Lines 791–791:* **Still to write in Part 3, toward the 130 target — 60 cards:** the argument and its stages (the escalation ladder, tactic change inside an argument, the apology that is not one, negotiation and the walk-away, the fight that is really about something else); first meetings and introductions (the interview, the date, the doorstep, the reunion after years); dialogue under physical constraint (whispered, shouted across distance, spoken while working, spoken in the dark, spoken in a language of gesture); dialogue with non-human and non-present interlocutors (the prayer, the letter read aloud, talking to the dead, talking to a machine, the AI interlocutor, the animal); children's and preschool dialogue; period and heritage speech (how much to archaise, the anachronism budget, translated-from-antiquity convention); dialogue for animation and for game branching; and the passes — the dialogue polish pass, the read-aloud pass, the cut-every-third-line pass, the parenthetical audit, and the tag-and-beat clean-up. --- ## `phase-02-screenplay-dialogue/cat-dial.md` *Lines 555–555:* **Still to write in this category:** voice and idiolect (signature vocabulary, verbal tics, register shift, second-language rhythm, professional argot), comedy of dialogue (the literal answer, escalating agreement, the non sequitur, the undercut, comic rule of three), monologue craft, group and ensemble scenes, telephone and mediated conversation, and translated or bilingual dialogue. Target for the full category: 130. --- ## `phase-02-screenplay-dialogue/cat-emo-2.md` *Lines 633–633:* **33 cards** in this file, across 6 sections: More of the specific emotions (11), Extremes and aftermath (4), Emotion at length (4), Performance, the page and the song (3), Commissioned and interactive work (5), Failure modes (6). *Lines 635–635:* Running total for category `emo`: **60 of 60** — 27 in `cat-emo.md`, 33 here. *Lines 639–639:* **Genuinely outstanding, named honestly.** Three topics were identified in research and cut for space rather than for lack of substance, and belong in a Part 3 if the category is extended past 60: **writing emotion for a child performer** — what a five-year-old can and cannot deliver, and how that constrains the line on the page; **tenderness as distinct from love** — care directed at the vulnerable, which behaves differently from `emo.love_without_declaration`; and **the emotional reset between episodes** in procedurals and preschool series, where a baseline must be restored without the restoration reading as amnesia. Three cards. If they are written, the target moves to 63. *Lines 641–641:* Three cards here carry a review date: `emo.episode_emotional_shape`, `emo.song_emotional_beat`, `emo.brand_emotion`. The other 30 are durable. --- ## `phase-02-screenplay-dialogue/cat-emo-3.md` *Lines 231–231:* **12 cards** in this file. Running total for category `emo`: **72** — 27 + 33 + 12. *Lines 233–233:* Sections: Reassurance and bad news (3), Permission and company (3), The commissioned address (3), The three Part 2 named as cut (3). *Lines 243–243:* Third, **the Permission intent needed three cards rather than one.** `intents.md` asked for the card that names the intent, and that is `emo.permission_piece`. But the two things a permission piece is built out of — the accurate sentence and the resembling account — are separately teachable and are each used far outside permission work. The brief's items 3, 4 and 5 are therefore correctly three cards, and they are kept so. *Lines 245–245:* **Still missing in this category, recorded honestly.** Two. There is no card for **the commissioned tribute for a specific bereaved family** — the eulogy as a writing job rather than as a register, which `voice.eulogy_register` and `dial.the_eulogy` approach from the voice and dialogue sides but which neither owns. And there is nothing on **emotion under translation and subtitling** — which emotional registers survive being rendered and which invert — a real constraint in this database's own Arabic-subtitled use, touched only obliquely at `rhythm.translated_rhythm` and `theme.untranslatable_value`. Both are genuine cards. Neither is in these 12, because neither was a gap the router reported. --- ## `phase-02-screenplay-dialogue/cat-emo.md` *Lines 433–433:* **Still to write in this category:** fear and its varieties, joy and why it is hardest to write, envy, loneliness, humiliation in comedy specifically, emotion in kids' and preschool content, emotion in advertising and brand work, emotion across a series arc, catharsis and its structural requirements, emotional beats in music video and lyric, writing emotion for performance versus for the page, emotion in interactive work where the audience acts, and cultural registers of emotional display. Target for the full category: 60. --- ## `phase-02-screenplay-dialogue/cat-pov-2.md` *Lines 434–434:* **23 cards** in this file, across 7 sections: Tense, Frame and the Layers of Telling (4), Who the Narrator Is Talking To (3), Difficult Owners (4), POV Under Structural Pressure (4), POV in Other Deliveries (5), The Workbench (2), The Failure Modes (1). *Lines 438–438:* Running total for category `pov`: **55 of 55** — 32 in `cat-pov.md`, 23 here. *Lines 440–440:* One genuine gap remains, recorded rather than padded: **the borrowed voice** — a viewpoint assembled from other fiction rather than from people — is described in prose inside `pov.divergent_cognition` and `pov.positionality` but has no card of its own, because writing it would have pushed this file to 24 and broken the declared count. It is a real named failure and the category target should move to 56 when it is written. --- ## `phase-02-screenplay-dialogue/cat-pov.md` *Lines 655–655:* **32 cards in this file.** Target for the `pov` category is 55, so **23 remain for Part 2.** --- ## `phase-02-screenplay-dialogue/cat-rhythm-2.md` *Lines 863–863:* **Running total for category `rhythm`: 72 cards (32 in Part 1 + 40 here). Three slots remain against the stated 75 target.** *Lines 865–865:* Still uncovered, and honestly more than three cards' worth — Part 3 should either take the three highest-value items or the category target should widen to about 85: --- ## `phase-02-screenplay-dialogue/cat-rhythm-3.md` *Lines 310–310:* Running total for category `rhythm`: **82** — 32 in `cat-rhythm.md`, 40 in `cat-rhythm-2.md`, 10 here. Part 2 stated a 75-card target with three slots left and said honestly that more than three cards' worth remained. Ten were written, and the target moves to 82 rather than seven real techniques being cut to protect an estimate made before any card existed. Part 2's own outstanding list is a separate matter and is still outstanding — see below. *Lines 314–314:* *The brief's item 6 was the parent of items 1–3, not their sibling, so it is written first.* `rhythm.duration_locked_line` carries the arithmetic and the one rule that governs every case — write long, cut to the number, never pad — and the three specific cards inherit from it. Written last, it would have been a summary of cards the reader had already read. Written first, it is the card the other three point back to, and it is also the one that reaches formats none of the others touch: motion graphics, corporate, the timed caption. --- ## `phase-02-screenplay-dialogue/cat-rhythm.md` *Lines 614–614:* **Remaining toward the 75 target: 43 cards for Part 2.** Not yet covered: the periodic and the loose sentence as formal opposites; the balanced sentence and antithesis; the broken parallel and the fourth item that breaks the set; chiasmus; the semicolon; the ellipsis; parenthesis and the whispered aside; opening on a subordinate clause; syllable count and the monosyllabic close; assonance and vowel colour; accidental internal rhyme; repeating a single word across a page; the callback line; paragraph-length variation across a chapter; the section break and the chapter drop; rhythm in first person against third; rhythm in present tense; interiority rhythm against action rhythm; pace across a whole chapter and scene-sequel alternation; comic timing at sentence level; rhythm in headlines, decks and standfirsts; rhythm in email and chat register; lyric rhythm against prose rhythm; rhythm in Arabic and other right-to-left prose; read-aloud rhythm for picture books and children's writing; rhythm in instructional and technical writing; the quoted passage inside your own prose; typography, measure and line length as pacing; mobile reading against print; the audiobook paragraph; pace in serialised release; and the remaining named failures — the stacked prepositional phrase, the buried verb, throat-clearing openings, the comma-comma-comma sentence, over-punctuation, the everything-is-a-fragment voice, the false-profound short paragraph, accidental sing-song metre, and rhythm borrowed wholesale from a model author. --- ## `phase-02-screenplay-dialogue/cat-scene-2.md` *Lines 667–667:* **40 cards in Part 2.** Running total for category `scene`: **71 of a target 120** (31 in `cat-scene.md`, 40 here). *Lines 673–673:* **Still to write in this category (Parts 3 and 4, 49 cards):** scene construction for short-form and vertical specifically; the action scene and the fight as a value turn; the sex scene and the intimacy scene; the comedy scene's engine and the scene-length joke; scenes of violence and their aftermath; dream, memory and imagined scenes; the courtroom, the interrogation and other procedural shapes; the scene in prose versus the scene on screen; the chapter as a scene container; scene construction in animation and for children; the documentary scene and the interview as a scene; the musical number as a scene; scene construction for games and interactive work; the opening scene and closing scene of a whole work; and the remaining diagnostic passes — the scene inventory, the location audit, the who-is-in-every-scene audit and the repetition test. --- ## `phase-02-screenplay-dialogue/cat-scene-3.md` *Lines 294–294:* **14 cards** in this file. Running total for category `scene`: **85** — 31 + 40 + 14. *Lines 316–316:* **Still missing after this file.** `cat-scene-2.md` projected a 120-card target across four parts and listed 49 outstanding techniques. This file does not deliver them — it was commissioned to close the reported gaps and does that instead, so the honest position is that the category closes at **85** with the following still unwritten and still real: the action scene and the fight as a value turn; the sex and intimacy scene; the comedy scene's engine; violence and its aftermath; dream, memory and imagined scenes; the courtroom, interrogation and procedural shapes; the scene in prose against the scene on screen; the chapter as a scene container; animation and children's scene construction; the musical number; interactive and game scenes; short-form and vertical scene construction; the opening and closing scene of a whole work; and the four diagnostic passes — scene inventory, location audit, who-is-in-every-scene, and the repetition test. --- ## `phase-02-screenplay-dialogue/cat-scene.md` *Lines 508–508:* **Still to write in this category:** scene openings and closings by type, the montage and its dramatic rules, crosscut and parallel scenes, silent scenes, group and ensemble scenes, the scene without conflict (Kishōtenketsu development), scene length and its relationship to value size, and the sequence-level joins. Target for the full category: 120. --- ## `phase-02-screenplay-dialogue/cat-sub-2.md` *Lines 500–500:* **24 cards** in this file, across 6 sections: More than two in the room (3), Channels that are not dialogue (5), Choice, code and name (4), The questioned subject (2), Subtext across a whole work (5), Handing it over and giving it up (5). *Lines 502–502:* Running total for category `sub`: **55 of 55** — 31 in `cat-sub.md`, 24 here. Every item named in Part 1's Coverage list has a card. Two honest notes on the mapping. First, the triangle bullet held two moves — the shifting alliance and the person who is never addressed — and they are written as one card, `sub.triangle_scene`, because they are the same piece of scene geometry read from two positions; splitting them produced two half-cards in draft. Second, the setting-and-weather bullet resolved into a failure rather than a technique: the positive use of a physical environment as the unspoken party is already held by `sub.object_as_confession` here and by `tens.corrupting_the_mundane` and `emo.object_carrier` elsewhere, and what this category uniquely owns is the sympathetic-weather failure, so `sub.weather_does_the_feeling` carries the corrective in its Avoid-when in the same way Part 1's failure cards do. *Lines 504–504:* One real area remains unwritten and the target may move for it: subtext in translation and in dubbing — what an indirection does when it crosses into a language whose politeness system differs, and what a dubbing writer is allowed to make explicit to fit a mouth. Part 1's `sub.across_languages` covers a character operating in a second language, which is a different problem. If that is written, `sub` becomes 56. --- ## `phase-02-screenplay-dialogue/cat-sub.md` *Lines 702–702:* Toward a 55-card target, 24 remain for Part 2. Not yet covered: subtext in monologue and direct address; subtext between more than two characters (triangles, alliances, the person in the room who is not spoken to); subtext carried by setting and weather; subtext in voiceover and narration; subtext in written documents inside a story (letters, texts, emails); music and score as subtext; subtext in action and fight scenes; the subtext of what a character chooses to eat, wear or buy; sexual subtext and its codes; subtext under interrogation and cross-examination; subtext in group comedy; the subtext of names and forms of address; subtext in genre-coded work (horror, sci-fi allegory); subtext that shifts on a second reading; the relationship between subtext and theme; subtext in adaptation; subtext in improvised and devised work; the actor's contribution and where the writer should leave room; subtext in documentary and interview; the difference between subtext and foreshadowing; running subtext across a series; subtext in short forms under sixty seconds; the reveal that converts subtext to text; and the retired subtext of a relationship that has finally spoken. --- ## `phase-02-screenplay-dialogue/cat-voice-2.md` *Lines 853–853:* Running total for the category: **72 cards** (32 in `cat-voice.md` + 40 here), against a 70-card target. Part 3 is therefore a refinement pass rather than a fill: it should take the items below and, if the category is to be held near target, the split should be rebalanced at build time rather than padded. --- ## `phase-02-screenplay-dialogue/cat-voice.md` *Lines 577–577:* Sections covered: the narrating persona (6), register and diction (6), authority and its instability (4), institutional / ceremonial / non-human voices (5), irony and understatement (3), dialect, translation and code-switching (4), voice in production (4). *Lines 579–585:* Toward the 70-card category target, roughly 38 remain for parts 2 and 3. Not yet covered: - **Persona and address:** the essayistic voice, direct address and the broken fourth wall, second-person narration as a voice choice, the interrogative voice, the naive voice as a sustained register, the voice of a diary against the voice of a letter. - **Register families not yet given cards:** the reportorial register, the polemic and the manifesto, the eulogy, the apology, the sermon, the legal register, sports-commentary register, the sales register and why audiences now hear it instantly, profanity as register, jargon as belonging, archaic pastiche and period-appropriate diction, anachronism as a voice fault. - **Comic registers:** sarcasm as distinct from irony, the exasperated voice, the over-explainer, comic repetition of a single phrase, the straight-man narrator. - **Authority:** borrowed authority (quoting a document or an expert), declared stake and stated bias, the voice that loses authority on purpose across an arc, the expert narrator who is wrong, authority and gender/class assumptions in narration. - **Production:** voice casting and how a written voice survives a performer, register in subtitles versus in dub, voice across platforms (long-form to caption), house voice against personal byline, AI-default register as a named defect, thesaurus voice and purple prose as named defects, the narrator who is smarter than the story. --- ## `phase-05-song-generation/cat-lyric-2.md` *Lines 779–779:* **33 cards** in this file, across 6 sections: Form, Continued (7), Rhyme and Sound, Continued (5), Prosody, Continued (5), More Song Types (7), Process and Craft (5), Further Named Failure Modes (4). *Lines 781–781:* Running total for category `lyric`: **65** — 32 in `cat-lyric.md`, 33 here. The estimate was 55; every card here answers an item Part 1's own Coverage list named, so the target moves to 65 rather than real craft being cut to fit an estimate made before any card existed. --- ## `phase-05-song-generation/cat-lyric-3.md` *Lines 289–289:* Running total for category `lyric`: **75** — 32 in `cat-lyric.md`, 33 in `cat-lyric-2.md`, 10 here. Parts 1 and 2 declared the category complete against its own Coverage lists, and they were: every card here answers a gap reported by a *format kit*, not by the category. `formats-music.md` names five of them explicitly in its own Coverage section, and four of those five are now written. This is what the database looks like when it audits itself honestly — the category was finished and still had holes, because the holes were visible only from the kits. --- ## `phase-05-song-generation/cat-lyric.md` *Lines 632–632:* **32 cards** in this part, covering: form and architecture (6), the hook and repeating parts (3), prosody and singability (4), rhyme (4), song types and their jobs (7), language, version and screen (4), and named failure modes (4). *Lines 634–641:* Toward the 55-card target for this category, Part 2 should add roughly 23 cards. The confirmed gaps, in priority order: - **Form:** verse-refrain (tag) form; the post-chorus and the "drop" lyric; the double chorus and the a cappella last chorus; the intro line and the cold open; the outro tag and the fade-out lyric; the reprise and how much it may change; the modulation moment and what the lyric must do there. - **Rhyme and sound:** the held or delayed rhyme; multi-syllable and compound rhyme; the rhyme-free lyric; alliteration and consonance as texture; the deliberately unstressed line ending. - **Prosody:** parallel syllable-count lock across verses; melisma and where a word may be stretched; syncopation and pushing a line ahead of the beat; writing for two voices and a duet split; harmony lines and the counter-lyric. - **Types:** the duet and the argument song; the protest song; the worship and congregational lyric; the comedy patter song; the jingle and the sub-fifteen-second hook; the diss and boast verse; the lullaby. - **Process and craft:** object writing and sense-detail gathering; the point-of-view choice (I / you / he) and when to switch it; the rewrite pass order; the co-write split; writing to a brief and to a client's title. - **Further failure modes:** the second verse that repeats the first; the pronoun that changes person by accident; the chorus that peaks before the bridge can go higher; the lyric that is all setup and never arrives. --- ## `phase-01-story-generation/formats-social.md` *Lines 973–973:* **"The platform-specific variants of the reel" is three kits, not one.** The existing **Reels, Shorts and TikTok** kit stays as the shared-craft entry — hook, body, payoff, one idea, sound-off — because that craft genuinely is common. But its own Constraints block already states that the three surfaces rank on different signals, and a kit whose ranking is the argument cannot make three arguments at once. TikTok ranks on saves, shares and comment activity and rewards a piece that leaves a gap; Instagram ranks on reach and profile visits and caps boostability at 90 seconds; Shorts ranks on click-through and average view duration and is the only one of the three with real search discovery. Those produce three different rankings and three different endings, so they are three kits sitting under the shared one. --- ## `phase-01-story-generation/intents.md` *Lines 869–869:* 40 intents. The target set in the first pass is met. --- ## `phase-01-story-generation/symptoms.md` *Lines 642–645:* - Cards in the database: **1868** - Distinct ids named by a route here: **1699** — the page names **1701** in all, the extra two being named in this block and nowhere else - Complaints held: **544** - Broken links: **0** — every id on this page resolves to a live card *Lines 647–650:* What is not routed by complaint, and why: - **95** situation entries, which `situations.md` routes by situation rather than by defect - **74** individual structure-system beats, reached instead through their parent system card and through `structures.md` *Lines 652–652:* Those 169 are the whole remainder, and they are the whole remainder by design rather than by omission. **Every other card in the database now has a way in from a complaint.** The eleven interactive non-fiction cards once listed here as the one open gap — `inter.archive_way_in`, `inter.balance_cast_testimony`, `inter.branch_sourcing`, `inter.clicked_therefore_understood`, `inter.path_that_omits`, `inter.symmetrical_fabrication`, `inter.testimony_selector`, `inter.unchosen_path`, `inter.user_asked_question`, `inter.visitor_arc` and `inter.walk_up_entry` — are routed from the Interactive and branching section above, each named beside the complaint it answers, as is the twelfth, `inter.consent_without_a_cut`, from "The film has become something else and I have to go back to them", and the five stragglers before them: `comedy.physical_on_page`, `comedy.rule_of_three`, `comedy.short_form_shape`, `time.serial_clock` and `voice.sentence_rhythm`. *Lines 662–662:* `cat-char-3.md` and `cat-tens-4.md` are routed in full, in 37 new complaints and 16 extended ones. The character batch covered the areas character design was thinnest in, and most of it went into the Character section where the writer will already be looking: voice signature as a countable specification rather than a catchphrase, bodies and conditions written as constraint sets, desire for another person as a second want the story can put on the losing side, the family role and the rank and the unit and the one face lifted out of a crowd, animals and machines and gods and personified places, twins and doubles and assumed identities, and the development tools. Two clusters sit slightly away from that. The **names** group — `char.transliteration_choice`, `char.decorative_exotic_name`, `char.honorific_system`, `char.kinship_address` and `char.changed_name` — was given four complaints of its own next to "Writing across a culture I am not from", because a writer working in Arabic, Syriac or any language with honorifics and kinship address arrives at it with a real question and this page had nothing on it until now. And the series-maintenance cards were attached to complaints that already existed rather than given new ones: `char.bible_entry` and `char.behavioural_continuity` to "Three other writers will have this character after me", `char.write_for_known_actor` to "I can hear the writer casting the part", `char.animal_character` and `char.machine_character` to "The dog / the ship / the AI is set dressing", `char.crowd_member` to "The crowd scene is just noise", `char.idiolect_sheet` and `char.never_says` to "Every character sounds like me", `char.degraded_decision` to "The character under pressure isn't convincing", and `char.wound_questionnaire` to "The want is vague", which is the only place it earns its keep. --- ## `phase-02-screenplay-dialogue/scenes-2.md` *Lines 590–590:* 25 scene types. Running total across both files: **61**. *Lines 592–592:* **The 11 from Part 1's outstanding list:** the birth, the heist briefing, the chase, the seduction, the proposal, the breakup, the confrontation with a parent, the hospital waiting room, the ceremony, the eviction, the audition. *Lines 594–594:* **The 14 added here, which the set lacked and no list had named:** the ultimatum, the drive, the search of the house, the offer, the performance, the wake, the overheard conversation, the escape, the identification, the verdict, the first day, the bribe, the packing, the arrival of the stranger. *Lines 596–596:* **`scenes.md`'s Still-to-write list should be struck.** Its remaining 14 names — the first meeting, the job interview, the deathbed, the trial testimony, the intervention, the return to a childhood home, the border crossing, the phone call between two locations, the scene in a second language, the scene with a child present, the crowd, the shift handover, the last day at work, and the scene where nothing is said — are all written in that file already. The list is stale, not outstanding. --- ## `phase-02-screenplay-dialogue/scenes.md` *Lines 805–805:* 36 scene types. *Lines 809–809:* **Continued in `scenes-2.md`** — 25 further scene types, bringing the set to 61. An earlier version of this line listed 25 outstanding scene types; 14 of them had already been written into this file by the time it closed, so that list was stale and is struck. `scenes-2.md` records what both files still lack, together with the card-shaped gaps its entries had to fill in prose.
SHA-256: 05ed51bab010c399b5715dc30a3f21a16a9f53ef7a74c01244dcda678760e7de