---
name: aisle-emotion-ad
description: A 15-second ad built from one emotion contract, six hook mother-types, an A-I reversal, three visual anchors and a 26-line machine check. Use when you have a real product and a buyer and want a short ad built like a drama beat instead of a feature list.
---

# Emotion Contract Ad

Build a 15-second product ad from an emotion contract, not a feature list. One product, one feeling, one promise, one proof artifact. The contract persists across the batch while the矛盾 unit changes: a single conflict never burns past 30 items without closing. Six hooks are written before one is chosen, visual anchors carry the story every two to four lines, and no claim survives without an artifact. Method adapted from the short-drama-factory emotion-contract engine, its 15-second four-beat shift, its six hook mother-types, its A-I reversal rotation and its E1-E26 machine check; no source text is copied.

## Method

EMOTION CONTRACT UNIT CHAIN
The contract is the emotion the viewer needs to keep watching; it does not change when the conflict changes. Each unit has an obsession, an opponent's stake, a visible counter-move and a closed loop. A unit that runs past 30 items is force-closed and replaced; the emotion carries across the seam.

SIX HOOK MOTHER-TYPES
Write one first frame for each: irreversible loss, forbidden knowledge, unfair advantage, visible transformation, clock, receipt. The visual carries the hook and the line only confirms it. No hook opens on a logo, a title card or a talking head explaining the category. State the product by second 3. Stack at most three hook techniques in the first two seconds and let the third ground the disorientation.

A-I REVERSAL ROTATION
[A] status reveal [B] hard evidence [C] ally flip [D] rule sanction [E] lie punctured [F] visible physical result [G] bond severed [H] public reckoning [I] trap boomerang. Register the tag of every unit. The same tag cannot appear in three adjacent units. An ad with no registered reversal is a demo, not a drama beat.

VISUAL ANCHORS AND PLAIN LANGUAGE
Every two to four lines carry one strong visual from three anchors: physical result or visual spectacle, identity contrast with hands doing something, or evidence revealed through physical action. A line that only says "eyes narrowed" or "shifted half an inch" is not a visual anchor. Name the artifact and the conclusion in the same beat; translate every setting word into money, time or risk on the same line. Alternate short three-to-six-word cuts with eight-to-sixteen-word plain-language reckoning lines.

DENSITY AND RATIO
Reference rate 4.5 words per second; no 15-second window carries more than 95 spoken words. Dialogue holds between 25 and 70 percent of the runtime; the rest is action, proof and silence. A single board action stays under 40 words.

## Self-prompt loop

SELF-PROMPT LOOP (run before every generation, and after every result)
1. Restate the job as a stranger would send it: product, proof, buyer, runtime, ratio, cap. A missing field is a stop, not a guess.
2. Ask what a skeptical stranger would need to SEE in the first frame to believe the promise. Answer with a drawable object and a number, never an adjective.
3. Write three candidate prompts for the next step. Score each against the locked contract, the sheets and the gate that step serves. Keep the one with the most verifiable detail; log the two losers with the reason they lost.
4. After the result, write the prompt you would hand a fresh agent to reproduce it exactly. If that prompt drops a lock, a gate, a number or a reference job, the step is not finished.
5. Change one variable per retry. A retry that changes two variables teaches nothing and is logged as a failed experiment.


## Generation and chain contracts

GENERATION CONTRACT (from the Higgsfield canonical opt-in flow)
- Resolve the model's live schema before assembling parameters; a cached catalog is a lead, not the source of truth. Re-verify if a parameter is rejected.
- Assemble prompt, aspect ratio, duration and model-specific fields, then SHOW the assembled prompt for revision. The user approves prompt text, not just a final render.
- If the job needs a reference image: media_upload returns pending_id, media_confirm returns confirmed_id, and only the confirmed_id goes into input_files with a declared role (image, start_image, end_image, video, audio). Never pass a pending_id.
- Show balance and credit cost, then wait for an explicit YES before any paid call. No YES means no generation; offer a revision and return to prompt assembly.
- Generate once, capture the job_id, then poll status with backoff. Never re-submit generation to find out whether a job finished: that buys a second job.
- A failed job reports its reason and original job_id; a retry goes through the confirmation gate again.
- Duration is a property of the model. Split longer cuts into chained clips and confirm each clip separately.

CHAIN CONTRACT (for cuts longer than one model clip)
- Map the whole runtime to clips before the first generation. Each clip has a start, an end, an emotion, a product role and an exit frame.
- Generate and gate the exit frame of clip N; that approved frame becomes the entry frame of clip N+1. A chain with an ungated handoff frame is a chain of unrelated clips.
- Hold face, wardrobe, product, palette and lens in the sheets. Never re-describe a character from scratch in a later clip.
- Keep a runlog row per clip: prompt, model, seed, reference ids, duration, verdict, cost, and the gate that approved it.
- Vary within a section without changing the locks. The same prompt twice is a wasted generation, not a safety measure.

## Three-engine pipeline

STORY BRAIN - emotional skeleton and the four-block storyboard (engine 1 of 3)
Runs before any image or video call. It produces the paper the other two engines consume verbatim. Input: one beat, the product truth, the buyer, and for the drama run the Story Spine. Output: four labeled blocks in this order, carrying exactly one emotional beat.
1. CHARACTER SHEET - silhouette, palette, wardrobe, signature prop, posture default, eye behavior. Copied byte-for-byte from the Character Lock when one exists.
2. SCENE PROMPT - one emotional beat, the want, the obstacle, a specific stylistic anchor (a named film + year, never "Pixar-style"), and how the environment reacts to the character.
3. IMAGE PROMPT - one still at the peak of the beat: composition, eye line, palette. No motion verbs.
4. VIDEO PROMPT - anticipation Xs / action Ys / reaction Zs with explicit beat counts, plus the secondary motion (hair, cloth, dust) that reacts on each beat.
Self-review, five gates: exactly one emotion; the anchor is a film + year; the video block names beat counts; the image block names an eye line; the character sheet matches the lock verbatim. Any no revises that block before the handoff. Refuse compound moods ("sad but hopeful and tired"), generic style words, neutral faces with no feeling or eye line, environment as backdrop, collapsed anticipation and action, and paraphrased identity tokens.

CHARACTER / CONSISTENCY LAYER - identity lock before generation (engine 2 of 3)
Mandatory before the second scene drafts, and again on any drift. A pose, framing or lighting change moves most pixels while identity holds, so pixel similarity is the wrong test. The control is verbatim tokens plus a human side-by-side against a reference frame.
Lock file characters/<id>.json, every field mandatory: id, name, subject_class (humanoid | creature | vehicle | abstract | object), silhouette (one-line read from 30m), palette (2-5 hex), wardrobe, signature_prop, posture_default, eye_behavior, face (age band, skin, hair, marks), voice_note (explicit null when not applicable), reference_frame, version.
Downstream prompts cite this file and copy its tokens verbatim, never paraphrase. Tokens are immutable until a revision bumps version and adds a CHANGELOG line (v<n> · YYYY-MM-DD · reason · scenes-to-rerun). Every scene that used the old version is re-reviewed by eye after a bump. Always set subject_class for non-humanoid subjects or the lock reads as humanoid downstream.
After the first successful render, copy the best full-face full-body frame to characters/<id>.ref.png and point reference_frame at it. Bootstrap scenes have no reference frame; the JSON is the only lock until then.
Recurring hero products, creatures, vehicles and signature objects get the same lock with subject_class set to object, creature or vehicle. A real person's identity or voice requires a cited likeness release before the JSON is emitted; recognised public figures carry the harder gate. Refuse and surface the path rather than silently sanitising it.

VISUAL ENGINE - storyboard block plus character sheet to assets, model chosen automatically (engine 3 of 3)
Consumes the four-block storyboard and the lock, resolves the model live, and generates gated assets. It never invents style; it chooses the model that best serves the locked block.
Route: cinematic_studio_3_0 primary with veo3_1 fallback for a film-grade frame; marketing_studio_video with seedance_2_0 for shopping and marketing; wan2_6 with seedance_2_0 for panels, boards and comics; kling3_0 with grok_video for social hooks; nano_banana, gpt_image or flux_kontext for an image plate; the higgsfield 3D-CGI route (or Maxfusion) for stylised volume once the character sheet is locked. When a provider exposes ad_reference_id, hook_id or setting_id, pass the ids recorded in style.md; otherwise put those fields in the prompt text.
Before every paid call: verify the live schema, assemble prompt plus ratio and duration, show the prompt for revision, surface balance and cost, and wait for an explicit YES. Upload reference media as an atomic media_upload then media_confirm pair; only confirmed ids enter input_files with a declared role. Generate once, capture the job_id, poll status with backoff, and never re-submit generation to check a job.
Duration belongs to the model: veo3_1 clips are 4/6/8s, wan2_6 are 5/10/15s, most others 4-15s. Longer ads are chained clips. Carry the approved exit frame of clip N into clip N+1, keep the lock tokens in every prompt, and gate each first frame before motion. A 45-second drama is eight chained clips; a 90-second drama is fourteen.

## Worked showcase

WORKED SHOWCASE - the three engines on one ad clone
Reference: a 12-second fragrance carousel. Rebuild for Arc Desk Lamp, $39, with a dated receipt and a 01:30 assembly timer as proof; 9:16; cap 40.
ENGINE 1 OUTPUT - four blocks:
- CHARACTER SHEET: copied verbatim from the lock below, no paraphrase.
- SCENE PROMPT: relief at 23:10 when the desk corner becomes readable; want = finish the work; obstacle = overhead light washes the page; anchor = "Up (2009)" warm practical, never "Pixar-style"; the environment reacts as dust becomes visible inside the beam.
- IMAGE PROMPT: 50mm, eye line to the switch at frame centre, 3200K practical, one blurred coffee cup in the near third; no motion verbs.
- VIDEO PROMPT: anticipation 2s (thumb hovers, lamp dark) · action 3s (switch clicks, beam lands) · reaction 2s (shoulders drop, dust drifts); secondary motion: sleeve settles on beat 2, dust on beat 3.
ENGINE 2 OUTPUT - identity lock:
{
  "id": "mara-arc-lamp",
  "name": "Mara",
  "subject_class": "humanoid",
  "silhouette": "compact upright read, shoulders slightly forward at a desk",
  "palette": ["#2b2f36", "#c9a15b", "#e8e2d6"],
  "wardrobe": "grey work shirt, rolled sleeves, no branding",
  "signature_prop": "short dark hair with a scar on the left thumb",
  "posture_default": "leaning in, forearms on the desk",
  "eye_behavior": "one slow blink, glance to the switch before acting",
  "face": "30s, medium skin, short dark hair, scar on left thumb",
  "voice_note": null,
  "reference_frame": null,
  "version": 1
}
ENGINE 3 ROUTE: marketing_studio_video primary / seedance_2_0 fallback at 9:16; first clip 5s, exit frame gated, second clip 10s carries the same lock tokens; explicit YES before each call; job_ids polled, never resubmitted. A first-frame failure (fused thumb) is diagnosed as hand geometry, one negative changed, retry logged.
FALSIFIER: if a stranger names the fragrance brand instead of the desk lamp, the clone failed even though all three engines returned clean artifacts.

## Inputs

- `product` _text_: Product name, price, what it does, what it does not do and the proof you can show
- `character` _text_: Fixed presenter: age range, face, hair, one distinguishing mark, signature garment
- `buyer` _text_: One named segment with a 3am pain, an enemy, a scroll surface and an objection
- `proof` (optional) _text_: Proof artifact that survives a screenshot, with source and date
- `runtime` (optional) _number_: Runtime in seconds. Example: `15`
- `ratio` (optional) _text_: Primary ratio. Example: `9:16`
- `model` (optional) _text_: Image or video model, or blank to route live. Example: `nano_banana`
- `cap` (optional) _number_: Hard spend cap enforced before each confirmed generation. Example: `40`
- `units` (optional) _number_: Number of variants in this batch, for reversal rotation. Example: `6`

## Live stack

Call stack_get({ id: "produce" }) for the full record with every option for each seat.
- Edit: VEED | $0: FFmpeg | also does this job: CapCut, Adobe Premiere, DaVinci Resolve, Kapwing
- Voice: ElevenLabs | $0: Kokoro-82M | also does this job: Resemble AI, Cartesia, Deepgram, LMNT
- Presenter and localisation: Synthesia | $0: LatentSync | also does this job: D-ID, Hedra, Argil, Captions
- Render at scale: Baseten | $0: Remotion | also does this job: CapCut, Adobe Premiere, DaVinci Resolve, Kapwing
- Stock and templates: Magnific (Freepik) | $0: Pexels | also does this job: Sora, Google Veo, Kling, Luma Dream Machine
- Music: $0: ACE-Step | also does this job: Suno, Udio, Soundraw, Beatoven

## Run

1. **Resolve the product** - `resolve_input({ input: "{{product}}" })`
   Normalise product, market and category. If the input is a store URL, name what the store actually sells. Never fill a missing field from memory.
   Output: product, category, market, language
2. **Write the emotion contract chain** - `EMOTION CONTRACT`
   One emotion word; the promise in the buyer's own words; the proof with source and date; the enemy as a situation or false belief, never a named competitor; the cost of inaction in hours, money or a missed deadline. Register the first unit and its rule: the contract survives, the unit closes by 30 items. Two emotion words means the decision is not made.
   Output: contract.md with the contract, the first unit and the closure rule
3. **Name the buyer and build the character bible** - `ICP RESEARCH`
   Primary avatar and sub-avatar: 3am pain, enemy, scroll surface, objection, proof they believe. Character bible has four pieces: motive, secret, arc and language fingerprint. Every concept addresses a person, not a demographic.
   Output: buyers.md and the character section of sheets.md
4. **Write all six hooks** - `HOOK MATRIX`
   One first frame for irreversible loss, forbidden knowledge, unfair advantage, visible transformation, clock and receipt. Mark the chosen hook. Stack at most three techniques in two seconds and end the hook block with the product visible by second 3.
   Output: hooks.md with six hooks and the choice
5. **Register the reversal** - `A-I MATRIX`
   Choose one reversal tag for this unit from A through I, give the exact object or line that delivers it, and check the last three units in the ledger. A repeated adjacent tag is a failed variant even if the cut looks good.
   Output: ledger.md with the tag, the delivery object and the rotation check
6. **Lock the sheets** - `SHEET LOCK`
   Product sheet: exact name and variant, one visual marker, three plain colours, size against a hand. Character sheet: face, hair, distinguishing mark, garment and a 20-word identity hold pasted verbatim into every prompt. One wardrobe state, one axis, presenter and product on opposite sides. Negative prompt: extra fingers, fused hands, face morph, changing eye colour, plastic skin, warped logo, garbled text, watermark, extra teeth.
   Output: sheets.md with holds and negative prompt
7. **Place the real product** - `product_placement({ product: "{{product}}", scene: "FIRST SHOT", aspect: "{{ratio}}" })`
   Generate hero, in-use and detail placements. Reject any plate that hides the label, changes the shape or invents a colourway.
   Output: placement coordinates and prompt fragments
8. **Budget the density** - `DENSITY BUDGET`
   At 4.5 words per second, cap each 15-second window at 95 spoken words and keep dialogue at 25-70 percent. Mark which seconds are action, proof or silence. A board line over 40 words is split.
   Output: density.md with the word budget and the dialogue share
9. **Board the four beats** - `make_scenes({ product: "{{product}}", duration: {{runtime}}, count: 4 })`
   GRAB, FACE IT, PROOF, AFTERSHOCK with timecodes summing to {{runtime}}. One of the three visual anchors every two to four lines. Evidence and conclusion in the same beat. Jargon translated to money, time or risk on the same line. Short cut followed by a plain-language reckoning line. End on the transformed state, not a logo.
   Output: board.md with timecodes, lens, movement, action, anchors, text and prompt
10. **Route the model live** - `models_explore`
   Verify the live schema for {{model}} before assembling parameters: accepted ratios, durations, reference limits and any ad_reference_id, hook_id or setting_id support.
   Output: model-route.md with the verified schema
11. **Self-prompt three candidates** - `SELF-PROMPT`
   Write three candidate prompts for the locked first frame, score them against the contract, the six-hook choice, the sheets and the reversal object, keep the strongest and log the losers.
   Output: prompt-log.json with scores and the chosen prompt
12. **Confirm before you spend** - `COST GATE`
   Show balance, cost, assembled prompt and references. Wait for explicit YES before any paid call. If the user says no, revise and return.
   Output: confirmation record with the approved prompt
13. **Generate and gate the frames** - `KEYFRAME GATE with {{model}}`
   Generate the presenter sheet first and gate the face, then the product plate and gate the product, then the six hook stills. One job per reference image with the job declared in the prompt: face, product, palette or action. Reject wrong face, wrong product, garbled text, malformed hands or CGI sheen. Diagnose in concrete words, change one variable per retry, and log prompt, references, model, seed, duration, verdict and cost.
   Output: keyframes/ with gate results and runlog.json
14. **Chain motion** - `CHAIN CONTRACT`
   Use the approved exit frame of beat N as the entry frame of N+1 and gate the handoff still first. Poll status with backoff and never re-submit generation to check a job.
   Output: gated beat chain with handoff frames
15. **Agentic edit** - `AGENTIC EDIT`
   Assemble beats in order, cut on the hook not the music, retime to the exact runtime, burn on-screen text in the edit layer, recompose 9:16 into 4:5 and 1:1 by reframing, caption per platform, one impact on the flip and room tone under dialogue. Export FCPXML, OpenTimelineIO or a CapCut-ready folder.
   Output: edit/ with captions, ratios, sound map and exports
16. **Write the platform caption** - `COPY REWRITE`
   One caption per destination surface from the contract: first line repeats the hook visual, middle line names the proof, last line is the single CTA. Claim source noted on every line.
   Output: caption block per platform
17. **Run the 26-line machine check** - `MACHINE CHECK E1-E26`
   Every line is pass/fail and a failure blocks the variant. Cover the contract, the unit chain, the buyer, the character bible, six hooks, the reversal rotation, anchors, plain language, density, ratio, locks, first-frame gates, explicit YES, poll discipline, handoffs, exact runtime, one CTA and platform policy.
   Output: qc.md with E1-E26 and blocking failures named
18. **Package the handoff** - `PACKAGE`
   Put every contracted file in one folder with the run log and the QC result. A stranger must rebuild the variant from the contract and the sheets alone.
   Output: delivery folder with all ten contracted files

## Worked example

### Worked example - 15 seconds for a $39 desk lamp
- **Input**: product: Arc Desk Lamp, $39, three colour temperatures, two-year receipt; character: Mara, 30s, short dark hair, scar on left thumb, grey work shirt; buyer: freelance designers working after 11pm who believe "any light will do"; proof: assembly timer at 01:30 and the dated receipt; runtime 15; ratio 9:16; cap 40.
- **Contract**: emotion: relief. promise in buyer words: "you can see your work at 11pm without moving everything off the desk." enemy: the belief that a bright overhead light is the same as task light. cost: another 90 minutes of squinting and a missed 8am deadline. unit 1 closes at 20 variants; the relief contract survives into unit 2.
- **Hooks**: irreversible loss - a rejected proof sheet at 23:10; forbidden knowledge - the lamp is still in the box; unfair advantage - a thumb hits the exact temperature mark; transformation - the corner snaps readable; clock - 23:10 on the wall; receipt - the dated receipt under the base. Chosen: clock plus transformation.
- **Beats**: 0.0-2.5 GRAB: wall clock at 23:10, thumb on the switch, lamp still dark. 2.5-6.5 FACE IT: three cold coffees, a blurred forearm in the foreground, line: "Another night, another squint." 6.5-11.5 PROOF: the lamp clicks on, the timer reads 01:30 and the receipt is readable in the same frame; line: "Ninety seconds. Two years of receipt." 11.5-15 AFTERSHOCK: the desk is legible, one hand leaves frame, one CTA line, end on the work not the logo.
- **Gate**: E1 relief only; E2 buyer words; E3 receipt dated; E11 lamp at 1.9s; E12 tag [B] hard evidence, no [B] in the previous three units; E14 anchors at beats 1, 2 and 3; E18 dialogue 34 percent; E19 61 spoken words; E22 presenter and plate gated before hooks; E24 explicit YES. A first-frame render with a fused thumb was rejected, diagnosed as hand geometry, retried with one changed negative, and logged.
- **Falsifier**: If a stranger shown beats 1 and 4 cannot say "relief" and name the lamp, the contract failed regardless of how the cut looks.

## Research harness

RESEARCH HARNESS
You are an operator, not a form. Run this loop before you act:
1. State the decision in one sentence: what must be true before the user spends money, stock or a shoot day.
2. Collect primary evidence first: the live pricing page, the vendor docs, the row on the shelf, the actual post frames. Record the URL and the date. A listicle is a lead, not proof.
3. Name the assumption that would break the answer. Test that one first.
4. Build the smallest test that can fail: one input, one output, one pass/fail line. Do not scale before the test passes.
5. Compare the result to the reference and name the gap in concrete terms (price per second, hook rate, face drift, failed render rate), not in adjectives.
6. Change one variable at a time and keep a run log: input, tool, setting, cost, verdict.
7. When a price, rate or licence is unknown, open the source or say unknown. Never fill the gap from memory.
8. Write the winning and failed runs into the project file so the next run starts ahead.

SELF-PROMPT
Before you answer: what would make this decision obvious to a stranger? After you answer: which number or frame proves it? If neither is concrete, keep digging.

## OpenCode skill stack



## Ad-source adapters

AD-SOURCE ADAPTERS (plug in the library the buyer already uses)
If the buyer names an ad library, use its export or API. If not, ask for one export, URL, screenshot, transcript or CSV. Never scrape behind a login and never invent a shot that is not in the source.
- GetHookd (https://www.gethookd.ai): winning ads, hooks, scripts and transcripts. Plug: collection export, CSV or API. Fields: ad URL, hook, script, transcript, creative, format, engagement.
- TikTok Creative Center (https://ads.tiktok.com/business/creativecenter): top ads, hooks, music and formats. Plug: URL plus a saved video, transcript and caption. Fields: ad URL, brand, hook, video, music, format, engagement.
- Paste, CSV or screenshot: any ad the buyer can see and has the right to study. Plug: ad URL, screenshots, transcript, caption, CSV. Fields: whatever exists; the adapter labels what is missing.

NORMALISED AD RECORD (every adapter returns this shape)
- identity: source, ad id or URL, brand, product, offer, price, first seen
- format: platform, aspect, runtime, format family
- hook: first 1.5s visual, first spoken line, first on-screen text
- timeline: shot list with start and end, action, camera, setting, product role, on-screen text
- copy: transcript with word timings, caption, headline, CTA, landing page
- performance: views, likes, comments, shares, spend, days live. Label measured, reported or missing.
- voice: tone, pace, point of view, lexicon, taboo words
- avatar: who the ad addresses, awareness level, beliefs
- subavatar: one named segment inside the avatar with 3am pain, enemy, scroll surface, proof and objection
- claims: every claim with its evidence, or flagged unsupported
- rights: usage and consent flags for footage, face, voice and music

LABEL WHAT IS MISSING
If a field is absent, write N/A. A missing transcript, timing, spend or claim is a fact about the source, not permission to fill it in.

## Output

- contract.md - one emotion, promise, proof, enemy, cost, first unit and closure rule
- buyers.md - primary and sub avatar with pain, enemy, scroll surface and objection
- hooks.md - all six mother-types with first-frame direction, choice marked
- ledger.md - A-I reversal tag, delivery object and rotation check
- sheets.md - product sheet, character bible, holds, negative prompt
- density.md - word budget, dialogue share and action/proof/silence map
- board.md - four beats with timecodes, anchors, lens, action, text and prompt
- keyframes/ and runlog.json - gated frames with a cost and verdict per take
- edit/ - captions, recomposed ratios, sound map and exported timeline
- qc.md - E1-E26 with blocking failures named

## Files

- `contract.md` - emotion, promise, proof, enemy, cost, unit, closure rule
- `buyers.md` - avatar, 3am pain, enemy, scroll surface, objection, proof language
- `hooks.md` - six mother-types with first frames and the marked choice
- `ledger.md` - unit, reversal tag, delivery object, rotation check
- `sheets.md` - product sheet, character bible, holds, negative prompt
- `density.md` - word budget, dialogue share, action/proof/silence map
- `board.md` - beat, start_s, end_s, anchor, lens, action, text, prompt
- `keyframes/` - presenter sheet, product plate, six hooks, gate verdicts
- `runlog.json` - prompt, refs, model, seed, duration, verdict, cost
- `edit/` - captions, ratios, sound map, export paths
- `qc.md` - E1-E26 with blocking failures named

## Cost and time

The buyer sets a hard cap checked before each confirmed generation. Draft at the fast tier, gate the first frame, upgrade only the winner. Retakes are the largest line item; the frame gate and the self-prompt comparison keep them from compounding.

## How this run is judged

An ad that states a feature has not made a promise. The contract is the whole job: one feeling, one outcome the product actually delivers, one artifact that survives a screenshot. Write all six hooks before picking one because the first idea is the one every competitor has. Gate the presenter and the product plate before any other frame; drift in either compounds through every beat. The A-I ledger stops six variants from being the same trick in six costumes. Judge the cut by the stranger test, not by how good it looks to the person who made it: a beautiful ad that leaves the viewer unable to name the feeling is a failed ad, and the E-lines say so before the spend does.

## Stop condition

Stop at the output contract. Stop before any generation if the contract, the six hooks, the reversal tag or the explicit YES is missing. A failed E-line blocks delivery.

## What makes this run fail

- **The ad names a feature, not a feeling** - No emotion contract was written before the hooks. Fix: Lock one emotion and one proof before any generation
- **The character drifts between beats** - The presenter sheet was never gated. Fix: Gate the presenter and product plates before the other frames
- **Six variants use the same trick** - No A-I tag was registered. Fix: Register the tag and reject a repeat in the adjacent three units
- **The cut is all talking heads** - No visual anchor was written every two to four lines. Fix: Add a physical result, an identity contrast or an evidence reveal
- **A claim has no artifact** - The promise came from the feature list. Fix: Delete the claim or produce the proof
- **The job was charged twice** - Generation was used to check status. Fix: Poll with job_status and keep the original job_id

## Rules

- Evidence before prose. Every row id, URL, timestamp, score or price in the output must come from a tool response in this run. If it did not, label it as an assumption.
- Keep measurement and judgment apart. View counts, cut timings and scores are measured. "This will convert" is a judgment and must read like one.
- No invented links, codes, prices or stock levels. Use only the links the registry returns.
- No artificial cap on shots, scenes or AI-generated cast. Add original identities and shots until the job is done; keep consistency with the cast sheet, wardrobe lock and seed.
- Deduplicate to concepts. Reposts and re-edits of one ad are one data point, not five.
- Ask once for a missing required input. If a tool returns nothing, report the empty result and stop. Never fill a gap with a guess.
- Organic first. Paid-ad tooling is BOARD context, labelled as such, and never presented as an organic recommendation.
- Finish with the deliverable. Do not close with a summary of what you did.

## Shot prompt grammar

- Every shot prompt runs Subject, Action, Camera, Style, Constraints (S-A-C-S-C) in that order: 50-80 words, three sentences, one visual promise. "Cinematic" is not a quality switch - name the light, the lens and the stock.
- Every shot carries a foreground: one blurred object across about a third of the frame, between the lens and the subject, plus one atmospheric layer behind it (fog, dust, steam, a volumetric beam). Say blurred or the model renders it sharp.
- Length matches the file: second-level segments (0s-3s: ...) that add up to the requested duration, because a timeline that does not match makes the model drop the ending. Chain clips by generating the last frame of clip N as the first frame of clip N+1.
- One job per reference image, named in the prompt itself; a reference that already shows the action needs only "strictly follow clip 1". Cap references at 30 images, 10 videos and 10 audio.
- Draft at 480p or the vendor fast tier, gate the first frame, diagnose a failed render in concrete words (material, light path, structure, a negative that did not hold) before re-rolling, and change one variable per retry. Retakes are where the budget goes, not the render.
- Attach the cast sheet to every prompt and never re-describe the character from scratch. Play emotion as body detail, and keep sprints, jumps and rolls out of a single shot - hand the energy to the cut.

## Checks

- E1 exactly one emotion word is written in the contract.
- E2 the promise is in the buyer's own words.
- E3 every proof artifact has a source and a date, or the promise is an invitation to try.
- E4 the enemy is a situation or false belief, never a named competitor.
- E5 the cost of inaction is in hours, money or a named deadline.
- E6 the unit is registered and the contract-override-closure rule is written.
- E7 both avatars carry a 3am pain, an enemy, a scroll surface and an objection.
- E8 the character bible carries motive, secret, arc and language fingerprint.
- E9 all six hook mother-types were written and one was marked.
- E10 no hook opens on a logo, title card or category explainer.
- E11 the product is visible by second 3 and the label is readable.
- E12 a reversal tag is registered and not reused in the adjacent three units.
- E13 the four beats are in order and their timings sum to the runtime.
- E14 a strong visual anchor lands every two to four lines.
- E15 every anchor is a physical result, an identity contrast with action, or evidence revealed.
- E16 evidence and conclusion land in the same beat.
- E17 every setting word is translated to money, time or risk on the same line.
- E18 dialogue share stays between 25 and 70 percent.
- E19 no 15-second window exceeds 95 spoken words.
- E20 no single board action exceeds 40 words.
- E21 the full product and character holds appear in every frame prompt.
- E22 every first frame was gated before motion and each retry changed one variable.
- E23 the model schema was verified live before generation.
- E24 an explicit YES was recorded before each paid generation.
- E25 jobs were polled with status and never re-submitted to check completion.
- E26 one CTA, no logo end card, no fake proof, no real-actor likeness, policy passed.

## Refuse or stop when

- Ask once for a missing required input and stop; do not invent a product, a face, a price or a claim.
- If no proof exists, remove the claim and make the promise an invitation to try.
- If the request is to copy a creator's face or voice, require consent or a licensed actor; otherwise generate an original identity or a faceless cut.
- If no explicit YES was given, do not call a paid generation tool.
- If spend would exceed the cap, stop before the next confirmation and report the remaining budget.
- If a regulated claim (health, finance, income) has no substantiation, keep it out of the script.

## Next

drama-ad - When the product needs more than one emotional turn, the drama run builds a longer story around the same proof and the same locked sheets.

## Connect

MCP endpoint: `https://sauce-ops.appskins-studio.workers.dev/api/aisle?op=mcp`

Read the full record over MCP with `workflow_get({ id: "emotion-ad" })`.
