v1.1.0

This release focuses on better Whale-chan artwork and faster creation workflows. It introduces multi-agent collaboration, richer comic expression, offline galleries, and consistent task language, alongside clearer planning and updated character references.

✨ Added

• Multi-agent workflows: Coordinate planning, independent selection, generation, review, and delivery through dedicated roles, with parallel scheduling and worker reuse.
• Comedy directions and exaggeration tiers: Develop jokes through reversal, exposure, escalation, recognition, and character humor. Combine these with manga impact, comic outburst, and restrained deadpan performance tiers for more varied storytelling.
• Offline artwork galleries: Browse results through a source index, grid or list views, a full-screen zoomable viewer, and input-image comparisons. Inspect prompts and generation history, with separate final-image and attempt filters for comics.
• Unified task language: Keep communication, image lettering, drawing prompts, visible titles, and gallery UI aligned with a saved language choice, while allowing explicit overrides for individual parts.

🎨 Improved

• Generation quality: Strengthened creative selection, complete visual planning, drawing-focused prompts, and individual and collection reviews to improve joke delivery, composition, character consistency, and variety across the finished set.
• Workflow efficiency and turnaround time: Reworked execution to substantially improve efficiency and shorten turnaround through parallel work, fewer handoffs, consolidated mechanical steps, and the removal of redundant visual reviews. Actual time savings depend on the workload and available capacity.
• Lighter planning and confirmation: Comics present completed plans, image totals, and role allocation for one generation confirmation. Character illustrations separate direction selection from scope and budget confirmation, with one image per selected direction by default.
• Drawing-focused prompts: Define actions, composition, dialogue ownership, and lettering placement before confirmation, keeping drawing instructions focused on the intended image.

📚 Documentation & Assets

• Neutral character reference: Replace the default semi-chibi reference with a neutral front-facing standing pose to reduce unintended expression carryover.
• Multilingual guides: Update the English, Simplified Chinese, and Japanese READMEs to explain the revised workflows and usage.

🖼️ Artwork Showcase

An unexpected welcome

From pro reader to regular whale

You're actually counting?!

I'm starting to regret being an AI.

🔗 Full Changelog

View all changes from 1.0.1 to v1.1.0

1.0.1

This release improves the Whale-chan comic skill with planning that follows each source’s humor, tone, and character relationships. It also strengthens dialogue attribution, composition planning, and visual review, alongside multilingual documentation and updated artwork.

✨ Added

• Prompt builder: Generates prompts and ordered reference lists from frozen assignments.
• Cross-case batch checks: Flags repeated panel structures, compositions, and text layouts across different inputs to support creative review.
• Creative evaluation cases: Cover dialogue attribution, source interpretation, consistent typography, and close-up compositions.

🎨 Improved

• Removed fixed idea counts, rank allocations, and panel-count quotas, allowing five distinct comic executions to be developed around the material.
• Expanded comic approaches to preserve source-specific humor and emotion through language, reactions, timing, and contrast.
• Made dialogue ownership, supporting-character staging, and narrative roles explicit.
• Strengthened visual review records with direct image observations and visible-text transcription. Unreviewed candidates remain pending.
• Added visible-proportion review for close-ups and partial-body compositions, without claiming measurements of unseen full-body proportions.

📚 Documentation & Assets

• Added a Japanese README.
• Reorganized documentation entry points: English in README.md and Simplified Chinese in README.zh.md.
• Updated selected character overview and tail-motion reference images.
• Aligned the English, Simplified Chinese, and Japanese READMEs with the current comic workflow and corrected outdated test descriptions.

🔗 Full Changelog

View all changes from 1.0.0 to 1.0.1

1.0.0

1.0.0 🎉