Generate detailed mood board descriptions with colors, textures, furniture, and lighting for a given interior design style. Use when the user wants to visualize a design concept before rendering.
Create rich, detailed mood board descriptions that capture the essence of a design concept — usable as creative briefs or as input for image generation.
Before generating, check what data already exists to enrich the mood board.
!ls projects/${PROJECT_NAME}/style-config.yaml 2>/dev/null && echo "HAS_STYLE_CONFIG=true" || echo "HAS_STYLE_CONFIG=false"
!ls projects/${PROJECT_NAME}/references/preprocessed/semantic_analysis.json 2>/dev/null && echo "HAS_SEMANTIC=true" || echo "HAS_SEMANTIC=false"
!ls projects/${PROJECT_NAME}/brief.md 2>/dev/null && echo "HAS_BRIEF=true" || echo "HAS_BRIEF=false"
!ls projects/${PROJECT_NAME}/notes.md 2>/dev/null && echo "HAS_NOTES=true" || echo "HAS_NOTES=false"
HAS_STYLE_CONFIG=true?
└── Read style-config.yaml — use exact colors (hex), materials, lighting settings.
Do NOT invent colors/materials — use what's configured.
HAS_SEMANTIC=true?
└── Read semantic_analysis.json — use real room data (existing materials, colors, furniture).
Mood board should build ON TOP of what exists, not ignore it.
HAS_BRIEF=true?
└── Read brief.md — extract mood/atmosphere, client preferences, constraints.
HAS_NOTES=true?
└── Read notes.md — check for feedback on previous mood boards or renders to avoid repeating mistakes.
None of the above?
└── Standard mode — gather all inputs from user.
A mood board grounded in real project data is 10x more useful than a generic one.
Gather from user or $ARGUMENTS:
/style-guide)Generate a comprehensive mood board with these sections:
A 2-3 sentence evocative description of the overall design vision. Written in present tense, as if describing a completed space.
List 5-7 textures present in the space:
5-8 specific furniture items:
Brief evocative description engaging multiple senses:
Present as a structured document with clear sections. End with:
/generate-prompt [style] [room] to create AI prompts from this mood board/generate-prompt — the natural next step. Converts this mood board into provider-optimized prompts. Pass the key elements (materials, colors, furniture, lighting mood) as input./design-consult — should have been used before this skill. If the user jumps straight to mood board without a design brief, suggest running /design-consult first to clarify requirements./style-guide — reference when building the mood board. Use style-specific vocabulary to ensure the mood board accurately represents the chosen style./render — after generating prompts from this mood board, /render sends them to AI providers./refine — if a previous render didn't capture the mood board's intent, use /refine to adjust the prompt based on what went wrong.