Firestarter

Shipped desktop app that plans, draws, and letters a comic chapter, with every speech balloon left as editable text

REPOgithub.com/Anuj-Attri/prod-imagen-studio
STACKElectron / TypeScript · Python 3.12 (standard library only) · Docker · multi-model image routing
PERIOD2026
STATUSreleased, self-updating installers

Firestarter is a drawing app for comics and illustration with an agent panel. Describe a scene and it lays out the panels and writes the dialogue into them. Ask for an eight-page chapter and it plans the chapter, shows you the plan, then draws every page. Speech balloons and captions stay as editable text objects on the canvas, so a line can be rewritten without regenerating the art. It also produces coloring books, posters, greeting cards, and system diagrams, and exports a chapter as a single PDF.

Poster generated from a one-line brief. The title and body text are canvas text objects, not pixels.
Poster generated from a one-line brief. The title and body text are canvas text objects, not pixels.

Product engineering, not a demo

A page from the comic path: the pipeline plans the beat, renders the art in the project's style, and lays lettering on top as editable canvas text.
A page from the comic path: the pipeline plans the beat, renders the art in the project's style, and lays lettering on top as editable canvas text.

Model bake-off

Before choosing defaults, the same prompts were run through GPT-5-image, Gemini 3 Pro Image, and Gemini 2.5 Flash Image, and the outputs kept side by side under version control so the decision is auditable. Dialogue generation has its own benchmark runs, stored as paired JSON and Markdown reports.

Brand-mark bake-off: the same prompt through GPT-5-image, Gemini 3 Pro Image, and Gemini 2.5 Flash Image, kept side by side under version control.
Brand-mark bake-off: the same prompt through GPT-5-image, Gemini 3 Pro Image, and Gemini 2.5 Flash Image, kept side by side under version control.

PropertyValue
Codebase~12.5k lines, Electron/TypeScript front end, Python back end
PlatformsWindows, macOS arm64, macOS x64, self-updating
Backendsingle stdlib-only Python file, Dockerised; hosted or self-hosted
LicensePolyForm Noncommercial, commercial rights reserved