Scenario Playable Ads
Overview
A playable is a 35 to 45 s slice of a real game that runs inside an ad network's container. It has:
- one self-contained HTML file with no network requests;
- one verb the viewer understands within 2 s;
- a game that waits for the viewer and never plays itself;
- a mute button, always;
- the genre's signature payoff;
- a win or an honest near-fail, then an end card with an install button.
Input: a gameplay recording and/or screenshots of the user's game, plus a logo, store links, a
description and a genre pick when the user has them. When the studio already has art, 3D, sound or
music, the build uses those files as they are, generates only what is missing in their style, or
derives the rest from one key art (references/existing-assets.md).
Output: dist/<network>/index.html for nine networks and a local preview, the ZIP each ZIP
network takes, and a bundle of all of them. A build ships only when it passes the 19 base QA checks
plus the per-network checks.
What was tested. The nine-network packaging ran against fake containers in headless Chromium: 64
of 64 checks on a copy of one game. The scripts in this skill, with the first-tap guard, the Pangle
SDK placement and the input and sound holds added since, pass 67 of 67 checks on the template's
placeholder game in the same fake containers. That game also passed AppLovin's playable preview
(portrait) and Google's HTML5 validator on the App Campaigns profile (2 November 2026). Google's
Campaign Manager and Display & Video 360 policies reject the google ZIP, which needs a clickTag
there. Meta's playable preview took the meta ZIP, passed its spec checks and registered the CTA
tap. No other network's own tool was run. Say which tools were run when you hand a build over, and
run each network's own preview first (references/networks.md). Each
archetype guide comes from one build that passed the 19 base checks.
Requirements: a local shell, Python 3.12 or later with Pillow, numpy, scipy, fontTools and brotli;
Playwright for Python with Chromium; ffmpeg. Install with
pip install "pillow>=13,<14" "numpy>=2.5,<3" "scipy>=1.18,<2" "fonttools>=4.60,<5" "brotli>=1.1,<2" "playwright>=1.64,<2"
then playwright install chromium; check with ffmpeg -version. The scripts, runtime and template
sit in this skill's scripts/ and assets/ folders, so the agent needs file access to the skill.
Connection, uploads and the core loop: the scenario skill. Image edits: scenario-image. Sheets
and cut-outs: scenario-game-assets. SFX and music: scenario-audio. Image-to-3D: scenario-3d.
If a sibling skill named here is missing from your available skills, load it with skill_get
(Scenario MCP) or ask the user to install it (https://github.com/scenario-labs/skills#install);
unattended, proceed from tool schemas and flag the gap.
Quick reference
| Step |
What happens |
| 1. Intake |
intake/: recording and/or screenshots, logo, intake.json (store links, networks, genre) |
| 2. Analysis |
contact sheet, key frames, analysis/brief.md and analysis/design.md (beats, asset map) |
| 3. Assets |
plate, transparent sheets, HUD crops, logo, SFX, music in assets/raw/, or studio files |
| 4. Build |
src/game.js; assets/manifest.json + sounds in index.src.html; prep_assets.py |
| 5. Package |
build.py --networks all (the default is meta,google) |
| 6. QA |
qa.py --speed 2, then look at every frame it writes |
| 7. Report |
notes.md: job ids, the source frame of every asset, tuning, open questions |
SKILL=<path to this skill>; P=<project dir>
cp -Rn "$SKILL/assets/template/." "$P/" && mkdir -p "$P/runtime" && cp "$SKILL/assets/runtime/runtime.js" "$P/runtime/"
python3 "$P/make_placeholders.py" "$P" --font "$SKILL/assets/fonts/fredoka/Fredoka-Bold.woff2" # optional
python3 "$SKILL/scripts/asset_inventory.py" "$P" # only with the studio's own files: analysis/assets_inventory.md
python3 "$SKILL/scripts/prep_assets.py" "$P" # assets/raw and intake/ to assets/web/
python3 "$SKILL/scripts/build.py" "$P" --networks all
python3 "$SKILL/scripts/qa.py" "$P" --speed 2 # qa/report.json and qa/*.png
cp -Rn never overwrites what the project already has, such as intake/.
- Never edit the runtime or the scripts per game. Work around a gap in the game's own code and write
it in
notes.md.
- Replace the template's game section with your genre. Keep its structure: config, reset, wait,
play, result, input on pointerdown, draw,
qa() with progress, boot.
- The hint hand is a vector drawing unless the project adds
assets/web/hand.webp.
Start here
When the user has not yet said what they have (for example "how do I make a playable ad?"), ask
once, in one message, for everything the build needs, then wait. Do not generate anything yet.
- A gameplay recording of 30 s to 2 min and/or store screenshots. With neither, stop: the skill
rebuilds what it sees and never invents a mechanic.
- The game name, the logo file, the App Store and Google Play links, and one line on what the
player does.
- The ad networks they run, or "not sure" (the build covers nine).
- Whether the game is 3D with a free camera (see the decision points below).
- Any art, 3D, sound or music they already have for the game, and whether they own it for ads
(references/existing-assets.md).
Say in two lines what they will get: the game's core loop rebuilt as code from the recording, with
their own art and sound where they sent some and generated art and sound for the rest, never the
recording itself, so side features are simplified. Then watch the recording and ask the genre
question below; when they sent files, ask in the same message how to use each kind (as supplied,
complement or derive). Those are the only two rounds of questions the user needs to answer; the
Scenario team and project, when the scenario skill asks for them, go in the first round.
Two questions before generating anything
Answer both in analysis/brief.md.
Which genre is it? Watch the recording first (contact sheet and key frames). Then ask the
user one question with these options:
- the two genres from references/genres.md that fit best, each with a
one-line reason from the recording (what the finger does, the most satisfying moment);
- "I don't know, you choose": pick the best fit and say why in one line;
- "Other": ask what the player does and what makes it fun, and store it as
game_description.
Record the answer in intake.json. Classify from the recording, not the label, and state your
confidence. No guide, a hybrid, or a game too big for one verb: follow
references/unknown-genre.md. Never invent a mechanic the recording
does not show.
Why would someone install after playing? Plan the levers from
references/make-them-install.md: a first touch that pays off
at once, visible growth, one real choice, a tease from the recording, a cliffhanger ending with
the player's own state on the end card.
Decision points
- Genre and product model (hypercasual, hybrid-casual, casual, midcore) go into
intake.json.
The genre's row names its guide or the nearest one.
- Hook: one opening pattern while the game waits: a power spike, a near-win, a satisfying
motion, or a clear promise. One short line on a backing plate. Fail-bait is allowed only as the
player's own choice.
- Signature payoff: the moment that makes the genre (a screen wipe, a crowd clash, a legendary
merge), at 10 to 25 s of play, the biggest and loudest thing in the ad.
- Ending: a good player wins, a careless one ends one step short, an untouched viewer gets the
end card after 10 s. Cap play at 35 to 45 s after the first touch.
- Original IP: when the genre has a famous leader, list what the game must not share with it
(characters, piece shape, board and track silhouette, how ammo is shown, the name). Have a human
compare with the leader's store page. See
references/archetype-paint-train.md.
- Networks: ask which ones the user runs. Per-network contracts, confirmed facts and the
unconfirmed list: references/networks.md.
- 3D games: ask before generating. A fixed or side camera means 2.5D from frames; a free 3D
camera means considering three.js with an image-to-3D model. Unattended, build 2.5D and say so.
See references/3d-games.md.
- Orientation: portrait first. Re-slice a landscape level along the portrait axis instead of
letterboxing it, and collapse an on-screen pad to one finger (say it is drift and ask the user;
unattended, ship the one-finger verb and record the drift in
notes.md).
Archetype index
| Archetype |
Ad verb |
Guide |
| Survivor shooter |
tap to move, auto-fire, tap a card |
survivor |
| Crowd runner with math gates |
drag left and right |
crowd runner |
| Lane shooter with gates |
drag left and right, auto-fire |
lane shooter |
| Merge-2 |
drag an item onto its twin |
merge |
| Sort puzzle |
tap source, tap target |
sort |
| Conveyor color puzzle |
tap to dispatch a shooter |
paint train |
| Pin pull |
tap or swipe a pin |
pin pull |
| Idle arcade |
tap to move |
idle arcade |
| Tower defense |
tap a pad to build or upgrade |
tower defense |
| Platformer, as a portrait climb |
tap where to land, tap again in the air |
platformer |
| Racing, real 3D |
drag to steer, tap to boost |
racing |
A new archetype gets references/archetype-<name>.md with these sections: Controls, Beat sheet,
Tuning, QA bot pattern, Asset map with source frames, Mistakes seen. Then add its row here and mark
it built in genres.md.
Rules every playable follows
- Never autoplay. The game never makes a move for the player: no demo, no rescue autopilot, no
auto-picked card. The world may run where that is the genre (a crowd runs forward), but every
choice is the player's.
- Wait, alive, for the first touch. The gameplay clock is held, the target glows, and the hand
loops the exact gesture on a real good move.
qa() exposes progress, which stays 0 until the
first touch.
- Three endings: the goal, a time cap of 35 to 45 s of play, and a 10 s idle timeout. After a
start, an honest line shows for at most 1.5 s before the end card.
- Hint, never a move. The hand returns after 2.5 s idle and only points.
- Mute button, always. The runtime draws it and it cannot be turned off;
cfg.mute only moves
it. It never counts as the first touch.
- The first tap never opens the store on
applovin, unity and mraid. The runtime enforces
it.
- Calm sounds. Nothing repeats faster than about 3 times a second (
cfg.soundGap). Duck the
music around each payoff (PA.audio.duck) and boost quiet takes with boost_db, never plain
gain.
- Honest result text, computed from the real state: "SO CLOSE! 1 left!" only when one move
remains.
- A near-fail for bad play. Scripted luck rewards the player's own good moves, so careless play
ends just short.
- Replay calls
PA.clearTweens() before reset(), and the game waits again.
Asset sourcing, prompts and model discovery:
references/scenario-assets.md. Starting from the studio's own
files: references/existing-assets.md. Lessons and known runtime
gaps: references/build-lessons.md. Every QA check, build refusal and
the game contract: references/qa-checks.md.
Worked example: a sort puzzle from a 40 s recording
- The user drops
gameplay.mp4, a logo and store links in intake/. A 4 fps contact sheet shows
taps that pour liquid between jars, and a jar that pops when full. Ask the genre question with
sort-puzzle and match-tile; the user picks sort-puzzle.
analysis/design.md: the hook "Can you sort them all in one go?", a jar pop at about 12 s, a
near-fail with one jar left, and a tease of a locked level 25 on the end card.
- Upload three frames. Generate the plate (an image edit that removes jars, liquids and HUD), one
transparent sheet of jars and the mascot, five SFX and a 30 s music loop, each via
recommend,
model_schema_get, dry_run, model_run with wait=false, and jobs_wait. Key the user's
logo with "key": true.
- Write the solver and the game from references/archetype-sort.md.
Run
prep_assets.py, then build.py --networks all.
qa.py --speed 2 fails idle_never_autoplays: a demo pour moved progress before the first
touch. Remove it, let the hand loop the pour gesture instead, and rebuild.
- Open every
qa/story_*.png: a banner covers the jar it describes. Move it to the top band.
- Hand over the bundle ZIP with
notes.md, and say which networks' own tools were run
(references/networks.md) and that the rest were tested only in fake
containers.
Common mistakes
- An ad that plays itself. A demo takes the move away from the viewer.
- A dead waiting screen, which fails
hook_moves_without_input and looks broken.
- A hint on the wrong spot. The hand must point at a real good move from the solver or the level
script.
- Chattering sounds: a tick per coin at 10 to 25 a second. One sound per event group.
- Paying for sound variants. A rising or alternate version of a sound is the same take played at
another
rate (PA.audio.play(name, vol, rate)), not a new job. Generate one take per event
group, and a second only for a take you reject: the archetype tables' "2 takes each" records past
builds, not a target. dry_run each sound payload and state the sum before launching any
(unattended: launch only within a stated budget, else report and stop). A variant that must not
share its sibling's repeat limit gets its own manifest and sounds entry pointing at the same raw
file: a new name, never a new job.
- Hard-coded near-fail text, which lies to players who lost by a lot.
- Copying recording glitches or platform UI, or baked HUD numbers. Redraw numbers live.
- Cutting a sheet by grid math. Sprites cross the cells: use
"blob": n and check --blobs.
- Decorative buttons that look live. Make them work, or make them a CTA.
- Units or zeros the game never shows on the end card.
- An asset with no source frame in the asset map.
- Editing the studio's files in place, or generating a piece they supplied.
- Claiming a network passed when only
qa.py ran.
Files
- Scripts, run by the agent: scripts/asset_inventory.py (catalog the
studio's own files), scripts/prep_assets.py (crop, cut out, key,
compress, subset fonts), scripts/build.py (inline and package per network),
scripts/qa.py (Playwright QA in fake containers),
scripts/pack_glb.py (a GLB into a JS include for three.js).
- Runtime, copied into each project: assets/runtime/runtime.js
(
window.PA).
- Template game: index.src.html,
game.js, manifest.json,
intake.json, qa_bot.py,
make_placeholders.py.
- OFL fonts with their licenses: Fredoka
(OFL), Rajdhani
(OFL),
Lilita One
(OFL),
Orbitron (OFL).
1---2name: scenario-playable-ads3description: "Use when a mobile game's gameplay recording, screenshots or existing art and sound should become a playable ad: an HTML5 interactive ad or interactive end card for Meta, Google Ads App campaigns, AppLovin, Unity Ads, Mintegral, Moloco, Liftoff, TikTok Pangle or any MRAID network, with a hint hand, a CTA and a mute button, packaged per network and QA-tested in headless Chromium. Keywords: playable ad, HTML5 playable, MRAID, FbPlayableAd, ExitApi, end card, CTA, hypercasual ad."4license: MIT5metadata:6 source: https://skillmd.com/skills/scenario-labs/scenario-playable-ads7---89# Scenario Playable Ads1011## Overview1213A playable is a 35 to 45 s slice of a real game that runs inside an ad network's container. It has:1415- one self-contained HTML file with no network requests;16- one verb the viewer understands within 2 s;17- a game that waits for the viewer and never plays itself;18- a mute button, always;19- the genre's signature payoff;20- a win or an honest near-fail, then an end card with an install button.2122**Input:** a gameplay recording and/or screenshots of the user's game, plus a logo, store links, a23description and a genre pick when the user has them. When the studio already has art, 3D, sound or24music, the build uses those files as they are, generates only what is missing in their style, or25derives the rest from one key art ([references/existing-assets.md](references/existing-assets.md)).2627**Output:** `dist/<network>/index.html` for nine networks and a local preview, the ZIP each ZIP28network takes, and a bundle of all of them. A build ships only when it passes the 19 base QA checks29plus the per-network checks.3031**What was tested.** The nine-network packaging ran against fake containers in headless Chromium: 6432of 64 checks on a copy of one game. The scripts in this skill, with the first-tap guard, the Pangle33SDK placement and the input and sound holds added since, pass 67 of 67 checks on the template's34placeholder game in the same fake containers. That game also passed AppLovin's playable preview35(portrait) and Google's HTML5 validator on the App Campaigns profile (2 November 2026). Google's36Campaign Manager and Display & Video 360 policies reject the `google` ZIP, which needs a `clickTag`37there. Meta's playable preview took the `meta` ZIP, passed its spec checks and registered the CTA38tap. No other network's own tool was run. Say which tools were run when you hand a build over, and39run each network's own preview first ([references/networks.md](references/networks.md)). Each40archetype guide comes from one build that passed the 19 base checks.4142Requirements: a local shell, Python 3.12 or later with Pillow, numpy, scipy, fontTools and brotli;43Playwright for Python with Chromium; ffmpeg. Install with44`pip install "pillow>=13,<14" "numpy>=2.5,<3" "scipy>=1.18,<2" "fonttools>=4.60,<5" "brotli>=1.1,<2" "playwright>=1.64,<2"`45then `playwright install chromium`; check with `ffmpeg -version`. The scripts, runtime and template46sit in this skill's `scripts/` and `assets/` folders, so the agent needs file access to the skill.4748Connection, uploads and the core loop: the `scenario` skill. Image edits: `scenario-image`. Sheets49and cut-outs: `scenario-game-assets`. SFX and music: `scenario-audio`. Image-to-3D: `scenario-3d`.50If a sibling skill named here is missing from your available skills, load it with `skill_get`51(Scenario MCP) or ask the user to install it (https://github.com/scenario-labs/skills#install);52unattended, proceed from tool schemas and flag the gap.5354## Quick reference5556| Step | What happens |57| ----------- | ------------------------------------------------------------------------------------------- |58| 1. Intake | `intake/`: recording and/or screenshots, logo, `intake.json` (store links, networks, genre) |59| 2. Analysis | contact sheet, key frames, `analysis/brief.md` and `analysis/design.md` (beats, asset map) |60| 3. Assets | plate, transparent sheets, HUD crops, logo, SFX, music in `assets/raw/`, or studio files |61| 4. Build | `src/game.js`; `assets/manifest.json` + `sounds` in `index.src.html`; `prep_assets.py` |62| 5. Package | `build.py --networks all` (the default is `meta,google`) |63| 6. QA | `qa.py --speed 2`, then look at every frame it writes |64| 7. Report | `notes.md`: job ids, the source frame of every asset, tuning, open questions |6566```bash67SKILL=<path to this skill>; P=<project dir>68cp -Rn "$SKILL/assets/template/." "$P/" && mkdir -p "$P/runtime" && cp "$SKILL/assets/runtime/runtime.js" "$P/runtime/"69python3 "$P/make_placeholders.py" "$P" --font "$SKILL/assets/fonts/fredoka/Fredoka-Bold.woff2" # optional70python3 "$SKILL/scripts/asset_inventory.py" "$P" # only with the studio's own files: analysis/assets_inventory.md71python3 "$SKILL/scripts/prep_assets.py" "$P" # assets/raw and intake/ to assets/web/72python3 "$SKILL/scripts/build.py" "$P" --networks all73python3 "$SKILL/scripts/qa.py" "$P" --speed 2 # qa/report.json and qa/*.png74```7576- `cp -Rn` never overwrites what the project already has, such as `intake/`.77- Never edit the runtime or the scripts per game. Work around a gap in the game's own code and write78 it in `notes.md`.79- Replace the template's game section with your genre. Keep its structure: config, reset, wait,80 play, result, input on pointerdown, draw, `qa()` with `progress`, boot.81- The hint hand is a vector drawing unless the project adds `assets/web/hand.webp`.8283## Start here8485When the user has not yet said what they have (for example "how do I make a playable ad?"), ask86once, in one message, for everything the build needs, then wait. Do not generate anything yet.87881. A gameplay recording of 30 s to 2 min and/or store screenshots. With neither, stop: the skill89 rebuilds what it sees and never invents a mechanic.902. The game name, the logo file, the App Store and Google Play links, and one line on what the91 player does.923. The ad networks they run, or "not sure" (the build covers nine).934. Whether the game is 3D with a free camera (see the decision points below).945. Any art, 3D, sound or music they already have for the game, and whether they own it for ads95 ([references/existing-assets.md](references/existing-assets.md)).9697Say in two lines what they will get: the game's core loop rebuilt as code from the recording, with98their own art and sound where they sent some and generated art and sound for the rest, never the99recording itself, so side features are simplified. Then watch the recording and ask the genre100question below; when they sent files, ask in the same message how to use each kind (as supplied,101complement or derive). Those are the only two rounds of questions the user needs to answer; the102Scenario team and project, when the `scenario` skill asks for them, go in the first round.103104## Two questions before generating anything105106Answer both in `analysis/brief.md`.1071081. **Which genre is it?** Watch the recording first (contact sheet and key frames). Then ask the109 user one question with these options:110 - the two genres from [references/genres.md](references/genres.md) that fit best, each with a111 one-line reason from the recording (what the finger does, the most satisfying moment);112 - "I don't know, you choose": pick the best fit and say why in one line;113 - "Other": ask what the player does and what makes it fun, and store it as `game_description`.114115 Record the answer in `intake.json`. Classify from the recording, not the label, and state your116 confidence. No guide, a hybrid, or a game too big for one verb: follow117 [references/unknown-genre.md](references/unknown-genre.md). Never invent a mechanic the recording118 does not show.1191202. **Why would someone install after playing?** Plan the levers from121 [references/make-them-install.md](references/make-them-install.md): a first touch that pays off122 at once, visible growth, one real choice, a tease from the recording, a cliffhanger ending with123 the player's own state on the end card.124125## Decision points1261271. **Genre and product model** (hypercasual, hybrid-casual, casual, midcore) go into `intake.json`.128 The genre's row names its guide or the nearest one.1292. **Hook:** one opening pattern while the game waits: a power spike, a near-win, a satisfying130 motion, or a clear promise. One short line on a backing plate. Fail-bait is allowed only as the131 player's own choice.1323. **Signature payoff:** the moment that makes the genre (a screen wipe, a crowd clash, a legendary133 merge), at 10 to 25 s of play, the biggest and loudest thing in the ad.1344. **Ending:** a good player wins, a careless one ends one step short, an untouched viewer gets the135 end card after 10 s. Cap play at 35 to 45 s after the first touch.1365. **Original IP:** when the genre has a famous leader, list what the game must not share with it137 (characters, piece shape, board and track silhouette, how ammo is shown, the name). Have a human138 compare with the leader's store page. See139 [references/archetype-paint-train.md](references/archetype-paint-train.md).1406. **Networks:** ask which ones the user runs. Per-network contracts, confirmed facts and the141 unconfirmed list: [references/networks.md](references/networks.md).1427. **3D games:** ask before generating. A fixed or side camera means 2.5D from frames; a free 3D143 camera means considering three.js with an image-to-3D model. Unattended, build 2.5D and say so.144 See [references/3d-games.md](references/3d-games.md).1458. **Orientation:** portrait first. Re-slice a landscape level along the portrait axis instead of146 letterboxing it, and collapse an on-screen pad to one finger (say it is drift and ask the user;147 unattended, ship the one-finger verb and record the drift in `notes.md`).148149## Archetype index150151| Archetype | Ad verb | Guide |152| ------------------------------- | --------------------------------------- | ------------------------------------------------------ |153| Survivor shooter | tap to move, auto-fire, tap a card | [survivor](references/archetype-survivor.md) |154| Crowd runner with math gates | drag left and right | [crowd runner](references/archetype-crowd-runner.md) |155| Lane shooter with gates | drag left and right, auto-fire | [lane shooter](references/archetype-lane-shooter.md) |156| Merge-2 | drag an item onto its twin | [merge](references/archetype-merge.md) |157| Sort puzzle | tap source, tap target | [sort](references/archetype-sort.md) |158| Conveyor color puzzle | tap to dispatch a shooter | [paint train](references/archetype-paint-train.md) |159| Pin pull | tap or swipe a pin | [pin pull](references/archetype-pin-pull.md) |160| Idle arcade | tap to move | [idle arcade](references/archetype-idle-arcade.md) |161| Tower defense | tap a pad to build or upgrade | [tower defense](references/archetype-tower-defense.md) |162| Platformer, as a portrait climb | tap where to land, tap again in the air | [platformer](references/archetype-platformer.md) |163| Racing, real 3D | drag to steer, tap to boost | [racing](references/archetype-racing.md) |164165A new archetype gets `references/archetype-<name>.md` with these sections: Controls, Beat sheet,166Tuning, QA bot pattern, Asset map with source frames, Mistakes seen. Then add its row here and mark167it built in `genres.md`.168169## Rules every playable follows170171- **Never autoplay.** The game never makes a move for the player: no demo, no rescue autopilot, no172 auto-picked card. The world may run where that is the genre (a crowd runs forward), but every173 choice is the player's.174- **Wait, alive, for the first touch.** The gameplay clock is held, the target glows, and the hand175 loops the exact gesture on a real good move. `qa()` exposes `progress`, which stays 0 until the176 first touch.177- **Three endings:** the goal, a time cap of 35 to 45 s of play, and a 10 s idle timeout. After a178 start, an honest line shows for at most 1.5 s before the end card.179- **Hint, never a move.** The hand returns after 2.5 s idle and only points.180- **Mute button, always.** The runtime draws it and it cannot be turned off; `cfg.mute` only moves181 it. It never counts as the first touch.182- **The first tap never opens the store** on `applovin`, `unity` and `mraid`. The runtime enforces183 it.184- **Calm sounds.** Nothing repeats faster than about 3 times a second (`cfg.soundGap`). Duck the185 music around each payoff (`PA.audio.duck`) and boost quiet takes with `boost_db`, never plain186 gain.187- **Honest result text,** computed from the real state: "SO CLOSE! 1 left!" only when one move188 remains.189- **A near-fail for bad play.** Scripted luck rewards the player's own good moves, so careless play190 ends just short.191- **Replay** calls `PA.clearTweens()` before `reset()`, and the game waits again.192193Asset sourcing, prompts and model discovery:194[references/scenario-assets.md](references/scenario-assets.md). Starting from the studio's own195files: [references/existing-assets.md](references/existing-assets.md). Lessons and known runtime196gaps: [references/build-lessons.md](references/build-lessons.md). Every QA check, build refusal and197the game contract: [references/qa-checks.md](references/qa-checks.md).198199## Worked example: a sort puzzle from a 40 s recording2002011. The user drops `gameplay.mp4`, a logo and store links in `intake/`. A 4 fps contact sheet shows202 taps that pour liquid between jars, and a jar that pops when full. Ask the genre question with203 `sort-puzzle` and `match-tile`; the user picks `sort-puzzle`.2042. `analysis/design.md`: the hook "Can you sort them all in one go?", a jar pop at about 12 s, a205 near-fail with one jar left, and a tease of a locked level 25 on the end card.2063. Upload three frames. Generate the plate (an image edit that removes jars, liquids and HUD), one207 transparent sheet of jars and the mascot, five SFX and a 30 s music loop, each via `recommend`,208 `model_schema_get`, `dry_run`, `model_run` with `wait=false`, and `jobs_wait`. Key the user's209 logo with `"key": true`.2104. Write the solver and the game from [references/archetype-sort.md](references/archetype-sort.md).211 Run `prep_assets.py`, then `build.py --networks all`.2125. `qa.py --speed 2` fails `idle_never_autoplays`: a demo pour moved `progress` before the first213 touch. Remove it, let the hand loop the pour gesture instead, and rebuild.2146. Open every `qa/story_*.png`: a banner covers the jar it describes. Move it to the top band.2157. Hand over the bundle ZIP with `notes.md`, and say which networks' own tools were run216 ([references/networks.md](references/networks.md)) and that the rest were tested only in fake217 containers.218219## Common mistakes220221- **An ad that plays itself.** A demo takes the move away from the viewer.222- **A dead waiting screen,** which fails `hook_moves_without_input` and looks broken.223- **A hint on the wrong spot.** The hand must point at a real good move from the solver or the level224 script.225- **Chattering sounds:** a tick per coin at 10 to 25 a second. One sound per event group.226- **Paying for sound variants.** A rising or alternate version of a sound is the same take played at227 another `rate` (`PA.audio.play(name, vol, rate)`), not a new job. Generate one take per event228 group, and a second only for a take you reject: the archetype tables' "2 takes each" records past229 builds, not a target. `dry_run` each sound payload and state the sum before launching any230 (unattended: launch only within a stated budget, else report and stop). A variant that must not231 share its sibling's repeat limit gets its own manifest and `sounds` entry pointing at the same raw232 file: a new name, never a new job.233- **Hard-coded near-fail text,** which lies to players who lost by a lot.234- **Copying recording glitches or platform UI,** or baked HUD numbers. Redraw numbers live.235- **Cutting a sheet by grid math.** Sprites cross the cells: use `"blob": n` and check `--blobs`.236- **Decorative buttons that look live.** Make them work, or make them a CTA.237- **Units or zeros the game never shows** on the end card.238- **An asset with no source frame** in the asset map.239- **Editing the studio's files in place,** or generating a piece they supplied.240- **Claiming a network passed** when only `qa.py` ran.241242## Files243244- Scripts, run by the agent: [scripts/asset_inventory.py](scripts/asset_inventory.py) (catalog the245 studio's own files), [scripts/prep_assets.py](scripts/prep_assets.py) (crop, cut out, key,246 compress, subset fonts), [scripts/build.py](scripts/build.py) (inline and package per network),247 [scripts/qa.py](scripts/qa.py) (Playwright QA in fake containers),248 [scripts/pack_glb.py](scripts/pack_glb.py) (a GLB into a JS include for three.js).249- Runtime, copied into each project: [assets/runtime/runtime.js](assets/runtime/runtime.js)250 (`window.PA`).251- Template game: [index.src.html](assets/template/src/index.src.html),252 [game.js](assets/template/src/game.js), [manifest.json](assets/template/assets/manifest.json),253 [intake.json](assets/template/intake/intake.json), [qa_bot.py](assets/template/qa_bot.py),254 [make_placeholders.py](assets/template/make_placeholders.py).255- OFL fonts with their licenses: [Fredoka](assets/fonts/fredoka/Fredoka-Bold.woff2)256 ([OFL](assets/fonts/fredoka/OFL.txt)), [Rajdhani](assets/fonts/rajdhani/Rajdhani-Bold.woff2)257 ([OFL](assets/fonts/rajdhani/OFL.txt)),258 [Lilita One](assets/fonts/lilita-one/LilitaOne-Regular.woff2)259 ([OFL](assets/fonts/lilita-one/OFL.txt)),260 [Orbitron](assets/fonts/orbitron/Orbitron-ExtraBold.woff2) ([OFL](assets/fonts/orbitron/OFL.txt)).