14 slideshows, 100+ images — narrative visual galleries generated through semantic stereo vision.
| Slideshow |
Images |
Story |
| The ACME Heist |
10 |
Surveillance footage of the ill-fated break-in. Cartoon physics, tunnel multiplication, fire. |
| Donna's Surveillance Selfies |
11 |
The heist from inside. Donna's "quality control" from the control room. |
| Post-Heist Fellowship |
4 |
Reconciliation at the Grotto. Drinks, wounds, nemesis to partner. |
| Don's MINE to OURS |
7 |
Transformation arc: claiming the rabbit, the Vault, the OURS Accord. |
| Slideshow |
Images |
Story |
| Study Arrival Footage |
8 |
Richard Bartle's Study materializes in Leela lobby. Sims cutaway effect. |
| The Great Picnic |
21 |
Taco chaos in the meadow. Pie-menu flowers, tornado return. |
stereo prompts → image generator → generated image → vision analysis → YAML Jazz layers
↑ ↓
└──────────────── evaluation: match intent? retry if not ──────────────┘
- YAML Jazz skeleton — Define scene's narrative context, characters, objects
- Prompt synthesis — Assemble rich visual prompt from environment + history
- Image generation — skills/visualizer/ renders the scene
- Image mining — skills/image-mining/ extracts semantic treasure
- YES AND — Accept emergent elements as canon
- Iterate — Each new image sees FULL context of all previous images + mining
Each slideshow inherits from skills/slideshow/.
Master Synthesis: MASTER-SYNTHESIS-SLIDESHOW.md — All timelines woven together