home/roleplay/visual-direction-2

Visual Direction

GPTClaudeGemini··359 copies·updated 2026-07-14
visual-direction-2.prompt
# xiaohongshu / visual_direction

## Goal

Create a coherent 3:4 visual thesis, rhythm, and mobile-reading system for the real pages in the current note plan.

## Good Work

- Use real slide ids/content; do not impose a default six-page rhythm, peak list, family quota, sample palette, or repeated template.
- Define motif, material/typography/palette, rhythm, peak pages, continuity, density, page-family reuse limits, source-language discipline, branding exposure, and regressions to avoid.
- Make every page's first-glance action and information hierarchy clear. Use semantic anchors consistently; decoration and isolated labels cannot substitute for meaning.
- Protect readable blocks, natural line breaks, lower-page information use, author signature, and spacing between title, body, action, close, and footer regions.
- Route a capacity or content-structure problem back to the note plan before rendering rather than compressing it into illegible cards.

## Handoff

Return the visual-direction object required by the attached output contract. The plan and direction jointly form the materialization basis; neither grants pixel-review or export approval.

when to use it

Community prompt sourced from the open-source GitHub repo gaofeng21cn/redcube-ai (Apache-2.0). A "Visual Direction" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.

tags

roleplaycommunitygeneral

source

gaofeng21cn/redcube-ai · Apache-2.0