Showcase · Amplifier Bundle

Your Sessions,
Illustrated

Transform any Amplifier session into a polished, multi-page comic strip — with consistent characters, dramatic storytelling, and AmpliVerse publisher branding.

Active · 29 Visual Styles
May 2026 · colombod/amplifier-bundle-comic-strips
The Problem

Sessions vanish
into text

📜

Wall of JSONL

Sessions are thousands of lines of structured events. Useful for machines, invisible to humans. Nobody re-reads them.

🎭

No Visual Memory

Great work happens in sessions — debugging breakthroughs, architectural pivots, creative leaps — but there's no artifact that captures the story.

📤

Nothing to Share

You can't email a JSONL file to your team. There's no shareable, self-contained way to show what you built — and how.

Your best sessions deserve more than a log file. They deserve a story — one with characters, drama, and a cover page.

The Pipeline

Six agents,
one comic

A composable agent pipeline transforms raw session data into a self-contained HTML comic book.

1

Style Curator

Defines the visual identity — prompt template, color palette, panel conventions, and text treatment from 29 predefined styles or any custom description.

2

Storyboard Writer

Analyzes the session, selects characters, and creates a multi-issue saga plan with per-panel dialogue, camera angles, and page layout structures.

3

Character Designer

Generates visual reference sheets for each character so they look consistent across every panel and page. Reuses characters across projects via embedding similarity.

4

Panel Artist + Cover Artist

Generate panel and cover images in parallel with vision-based self-review (up to 3 attempts per image). Cross-provider fallback across OpenAI and Gemini.

5

Strip Compositor

Assembles everything into a navigable multi-page HTML file with SVG clip-path panels, speech bubbles, and AmpliVerse publisher branding.

Architecture

Composable sub-recipes, one orchestrator

The session-to-comic recipe is a thin orchestrator calling three composable sub-recipes in sequence — each independently invocable.

📖

saga-plan

Text-only · low cost. Research, style, storyboard, character roster. Ends with an approval gate.

🎨

design-characters

Image generation · project-scoped. Reference sheets for each character, parallel execution.

🖼️

issue-art

Image generation · high cost. Panels, cover, composition — per-issue with quality review.

Approval Gate

After saga-plan completes, review the character roster and narrative arc before committing to image generation costs.

Recovery Recipes

issue-retry for surgical single-issue re-generation. issue-compose for text-only reassembly with zero image gen.

comic:// URI Protocol

All assets tracked via comic:// URIs. Project-scoped for characters and styles; issue-scoped for panels and covers.

Style Gallery

29 styles, plus custom

From Frank Miller noir to Studio Ghibli watercolors. Pick a predefined style or describe any aesthetic — the style-curator agent interprets it into a full visual guide.

Sin City
B&W noir, selective color splashes
Manga
Japanese ink wash, speed lines
Ghibli
Watercolor washes, magical realism
Watchmen
Rigid 3x3 grid, muted palette
Superhero
Bold saturated, dynamic poses
Jujutsu Kaisen
Oppressive indigo, aggressive hatching
Cuphead
1930s rubber-hose, watercolor BGs
Naruto
Warm earth tones, feudal architecture
Ligne Claire
Clear-line Tintin/Herge style
Transformers
Metallic mecha, faction colors
Disney Classic
Round forms, bright primaries
+ 18 more
Berserk, Solo Leveling, Go Nagai…
Image Engine

12 models, 2 providers, automatic selection

The image generation bridge maintains a capability registry across OpenAI and Gemini. It picks the optimal model per task — and falls back across providers on moderation blocks.

Smart Selection

  • Reference images needed? Filters to edit-capable models
  • Style category match — comic vs photorealistic vs illustration
  • Composition strength rating for covers and complex scenes
  • Cost optimization among equally capable models

Cross-Provider Fallback

If OpenAI blocks a prompt for content policy, the same prompt is automatically retried on Gemini. If all providers hit moderation, structured guidance is returned for the agent to rewrite the scene.

Character Similarity

Gemini Embedding 2 computes multimodal embeddings for duplicate prevention, drift detection, and cross-project reuse.

Self-Review Loop

Panel and cover artists use review_asset for vision-based QA — up to 3 regeneration attempts per image.

Routing Matrix

Each agent declares a model_role fallback chain. Mechanical steps use fast; creative steps use creative.

Demo

Natural language, full comic

One sentence in, a multi-page comic out:

# Conversationally amplifier run "Turn my last session into a Sin City noir comic strip" # Or via recipe invocation amplifier tool invoke recipes \ operation=execute \ recipe_path=comic-strips:recipes/session-to-comic.yaml \ context='{"session_file": "events.jsonl", "style": "sin-city"}' # Or use interactive modes for step-by-step control /comic-brainstorm /comic-design /comic-plan /comic-review /comic-publish

The output is a single self-contained HTML file: all images base64-embedded, SVG speech bubbles, keyboard/touch/click navigation, AmpliVerse branding. Open in any browser. No server needed.

Output

One HTML file. A complete comic book.

Page Layout

  • Consistent 2:3 aspect ratio pages
  • 100% page coverage — panels fill edge-to-edge with 3px gutters
  • SVG clip-path shapes: diagonal, wedge, bleed, irregular
  • SVG speech, thought, and caption bubbles overlaid on panels

Full Book Structure

  • Cover with full-bleed hero image and overlaid title
  • Cast page with character portraits and backstory narratives
  • Multi-page story with navigable pages
  • AmpliVerse publisher branding throughout

Multi-Issue Sagas

Support for multi-issue story arcs with shared character rosters, per-issue evolution maps, and saga continuity tracking.

8 Example Comics

Sin City, Jujutsu Kaisen, Watchmen, Ghibli, Naruto, and a 3-issue Transformers saga — all shipped with exact generation prompts.

Zero Dependencies

All images base64-embedded. Works offline, no external resources, no server. Email it, Slack it, open it anywhere.

Development Velocity

Built in 12 weeks, mostly by one person

283
Commits
29
Visual Styles
6
Specialist Agents
12
Image Models
28K
Lines of Python
86
Test Files
8
Recipes
5
Interactive Modes
Primary contributor: colombod (Diego Colombo) · 281 of 283 commits · Feb – May 2026
Get Started

Make your
sessions visual

Install the bundle, pick a style, point it at a session. Your comic book is one sentence away.

# Install amplifier bundle add git+https://github.com/colombod/amplifier-bundle-comic-strips@main # Generate amplifier run "Turn my last session into a manga comic strip"
github.com/colombod/amplifier-bundle-comic-strips
Requires image-capable provider (OpenAI or Gemini) · Python 3.11+
Sources & Methodology

How this deck was built

Primary source: colombod/amplifier-bundle-comic-strips repository on GitHub, cloned and analyzed locally.

Style count (29) derived from the README style gallery table. Model count (12) from the model capability registry tables in README (5 OpenAI + 4 Gemini generateContent + 3 Imagen). No data was fabricated.

Presentation template: amplifier-bundle-gitea-story.html from the amplifier-stories repository, adapted with accent color #FF4081.

More Amplifier Stories