Conversation
Major architecture changes based on extensive experimentation: 1. GENERATION: Replace 3-iteration generate-all loop with single front-loaded generation (all context + references in first message, simple "give me page N" requests per panel). Best character consistency across panels. 2. REVIEW LOOP: New review-and-reroll loop that distinguishes stochastic failures (re-roll) from systematic issues (fix prompt). Agent correctly identifies text leaks as stochastic and re-rolls instead of rewriting prompts. 3. CHARACTER REFERENCES: Inject curated character close-ups into the generation chat. Text specs alone cannot control visual consistency — visual references are essential. 4. PROMPT STANDARDIZATION: RENDERING RULES header on every prompt, standardized character specs with age/proportions/limbs/scale, font anchoring, stripped parenthetical stage directions. New scripts: generate-frontloaded-v2.ts, generate-panel-with-history.ts, diagnose-chat-context.ts New prompts: generate-panels-frontloaded.md, review-and-reroll.md See LEARNINGS.md for the complete play-by-play of all discoveries. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Major architecture changes based on extensive experimentation:
GENERATION: Replace 3-iteration generate-all loop with single front-loaded generation (all context + references in first message, simple "give me page N" requests per panel). Best character consistency across panels.
REVIEW LOOP: New review-and-reroll loop that distinguishes stochastic failures (re-roll) from systematic issues (fix prompt). Agent correctly identifies text leaks as stochastic and re-rolls instead of rewriting prompts.
CHARACTER REFERENCES: Inject curated character close-ups into the generation chat. Text specs alone cannot control visual consistency — visual references are essential.
PROMPT STANDARDIZATION: RENDERING RULES header on every prompt, standardized character specs with age/proportions/limbs/scale, font anchoring, stripped parenthetical stage directions.
New scripts: generate-frontloaded-v2.ts, generate-panel-with-history.ts, diagnose-chat-context.ts
New prompts: generate-panels-frontloaded.md, review-and-reroll.md
See LEARNINGS.md for the complete play-by-play of all discoveries.