emotional-canvas-moodboard

Name: emotional-canvas-moodboard
Author: Stanestane

34stars3forks

How to install emotional-canvas-moodboard?

stanestane/game-design-skills-bundle/emotional-canvas-moodboard

$npx -y skills add Stanestane/game-design-skills-bundle --skill emotional-canvas-moodboard

Installs into the current project.

Summary

Turn a game-design emotional canvas into a curated visual moodboard by deriving image-search queries, gathering a candidate pool from the internet, saving the files locally, rejecting weak references, and rendering both an HTML board and a flattened JPG board. Use when a user wan

Files · 1

View on GitHub

SKILL.md

1	# Emotional Canvas Moodboard
2
3	Translate a game's emotional direction into a concrete visual board.
4
5	Use this skill after `game-design-emotional-canvas` or whenever the user already has enough emotional direction to describe the intended feeling, adjacent feelings, visual tone, no-go elements, and reference-seeking targets.
6
7	## What to produce
8
9	Produce these artifacts unless the user asks for a smaller output:
10
11	1. A short emotional-canvas summary for the current concept
12	2. A search-plan list with focused image queries
13	3. A candidate pool of downloaded source images
14	4. A machine-readable `manifest.json` describing the curated board
15	5. A rendered `moodboard.html`
16	6. A rendered flattened `moodboard.jpg`
17	7. A short curation note explaining strong keeps and weak rejects
18
19	## Core rule
20
21	This skill is for reference gathering and board assembly.
22
23	Do not spend the whole turn writing abstract creative prose. Move quickly from emotional interpretation to query generation, candidate gathering, hard rejection, and board rendering.
24
25	A moodboard is curation, not accumulation. If the result looks like a pile of search results, the skill has failed.
26
27	## Workflow
28
29	### 1. Build or normalize the emotional canvas
30
31	If the user already supplied an emotional canvas, normalize it into this shape:
32
33	- Target feeling
34	- Three defining descriptors
35	- Adjacent feelings
36	- Contrasting feelings
37	- Visual references to seek
38	- Audio references to seek
39	- Narrative tone
40	- No-go elements
41
42	If the user only supplied a concept, create a compact emotional canvas first using the structure from `game-design-emotional-canvas`.
43
44	Keep it brief. The goal is to drive search quality, not to create a long essay.
45
46	### 2. Convert the canvas into image-search lanes
47
48	Create 6-12 focused search lanes. Each lane should represent one useful visual dimension, for example:
49
50	- shape language
51	- color and lighting
52	- environment mood
53	- texture and material feel
54	- composition and camera distance
55	- interface or arcade reference style
56	- character energy or motion language
57	- period/style anchors
58
59	For each lane, write 2-5 search phrases.
60
61	Good query pattern:
62	- `[subject/style] + [mood adjective] + [visual constraint]`
63
64	Examples:
65	- `arcade platformer mockup tense neon minimal background`
66	- `falling concrete blocks overhead cinematic top light`
67	- `retro industrial geometry oppressive playful contrast`
68	- `coins glowing on stacked blocks game art`
69
70	### 3. Gather a candidate pool, not a final board
71
72	Use web search and normal web access to find suitable visual references.
73
74	Prefer strong anchor sources over generic search sludge:
75	- official game store screenshots
76	- official key art
77	- ArtStation / Behance project images
78	- strong editorial or portfolio images
79
80	Avoid using random article headers, generic stock results, empty category pages, or filler photography unless they are genuinely the best aesthetic match.
81
82	Save source URLs in the manifest even when the file is downloaded locally.
83
84	Do not filter the collection by copyright status in this skill. The purpose here is fast local reference gathering. Copyright review can happen later in a separate workflow.
85
86	Build a candidate pool first:
87	- 16-40 candidates is normal
88	- mix broad atmosphere references with direct mechanic-adjacent references
89	- aggressively reject near-duplicates
90	- reject low-resolution thumbnails when a larger source is reachable
91
92	Quality bar for candidates:
93	- favor strong mood fit over literal mechanic fit
94	- favor visual authority over weak topical similarity
95	- prefer sources with clean composition and deliberate art direction
96
97	### 4. Curate brutally before rendering
98
99	Before rendering, judge each candidate on:
100	- aesthetic quality
101	- emotional fit
102	- readability / silhouette clarity
103	- color language
104	- geometric or material relevance
105	- uniqueness within the set
106
107	Then cut hard.
108
109	Final-board rule:
110	- default final board size is 6-12 images
111	- every image must earn its place
112	- each image should contribute a distinct job: atmosphere, shape language, color, space, movement, HUD/readability, collectible/reward language, material finish, or palette reference
113	- reject anything that is merely "related"
114	- prefer one strong image per anchor source/title over multiple samey screenshots unless a sequence is the point
115	- when the user wants a cleaner board, bias toward stronger composition and palette images even if they are less explanatory
116
117	Write down at least:
118	- why the strong

People who liked this also viewed

6 picks

Type

emotional-canvas-moodboard

34stars3forks

View on GitHub

How to install emotional-canvas-moodboard?

stanestane/game-design-skills-bundle/emotional-canvas-moodboard

$npx -y skills add Stanestane/game-design-skills-bundle --skill emotional-canvas-moodboard

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Emotional Canvas Moodboard
2
3	Translate a game's emotional direction into a concrete visual board.
4
5	Use this skill after `game-design-emotional-canvas` or whenever the user already has enough emotional direction to describe the intended feeling, adjacent feelings, visual tone, no-go elements, and reference-seeking targets.
6
7	## What to produce
8
9	Produce these artifacts unless the user asks for a smaller output:
10
11	1. A short emotional-canvas summary for the current concept
12	2. A search-plan list with focused image queries
13	3. A candidate pool of downloaded source images
14	4. A machine-readable `manifest.json` describing the curated board
15	5. A rendered `moodboard.html`
16	6. A rendered flattened `moodboard.jpg`
17	7. A short curation note explaining strong keeps and weak rejects
18
19	## Core rule
20
21	This skill is for reference gathering and board assembly.
22
23	Do not spend the whole turn writing abstract creative prose. Move quickly from emotional interpretation to query generation, candidate gathering, hard rejection, and board rendering.
24
25	A moodboard is curation, not accumulation. If the result looks like a pile of search results, the skill has failed.
26
27	## Workflow
28
29	### 1. Build or normalize the emotional canvas
30
31	If the user already supplied an emotional canvas, normalize it into this shape:
32
33	- Target feeling
34	- Three defining descriptors
35	- Adjacent feelings
36	- Contrasting feelings
37	- Visual references to seek
38	- Audio references to seek
39	- Narrative tone
40	- No-go elements
41
42	If the user only supplied a concept, create a compact emotional canvas first using the structure from `game-design-emotional-canvas`.
43
44	Keep it brief. The goal is to drive search quality, not to create a long essay.
45
46	### 2. Convert the canvas into image-search lanes
47
48	Create 6-12 focused search lanes. Each lane should represent one useful visual dimension, for example:
49
50	- shape language
51	- color and lighting
52	- environment mood
53	- texture and material feel
54	- composition and camera distance
55	- interface or arcade reference style
56	- character energy or motion language
57	- period/style anchors
58
59	For each lane, write 2-5 search phrases.
60
61	Good query pattern:
62	- `[subject/style] + [mood adjective] + [visual constraint]`
63
64	Examples:
65	- `arcade platformer mockup tense neon minimal background`
66	- `falling concrete blocks overhead cinematic top light`
67	- `retro industrial geometry oppressive playful contrast`
68	- `coins glowing on stacked blocks game art`
69
70	### 3. Gather a candidate pool, not a final board
71
72	Use web search and normal web access to find suitable visual references.
73
74	Prefer strong anchor sources over generic search sludge:
75	- official game store screenshots
76	- official key art
77	- ArtStation / Behance project images
78	- strong editorial or portfolio images
79
80	Avoid using random article headers, generic stock results, empty category pages, or filler photography unless they are genuinely the best aesthetic match.
81
82	Save source URLs in the manifest even when the file is downloaded locally.
83
84	Do not filter the collection by copyright status in this skill. The purpose here is fast local reference gathering. Copyright review can happen later in a separate workflow.
85
86	Build a candidate pool first:
87	- 16-40 candidates is normal
88	- mix broad atmosphere references with direct mechanic-adjacent references
89	- aggressively reject near-duplicates
90	- reject low-resolution thumbnails when a larger source is reachable
91
92	Quality bar for candidates:
93	- favor strong mood fit over literal mechanic fit
94	- favor visual authority over weak topical similarity
95	- prefer sources with clean composition and deliberate art direction
96
97	### 4. Curate brutally before rendering
98
99	Before rendering, judge each candidate on:
100	- aesthetic quality
101	- emotional fit
102	- readability / silhouette clarity
103	- color language
104	- geometric or material relevance
105	- uniqueness within the set
106
107	Then cut hard.
108
109	Final-board rule:
110	- default final board size is 6-12 images
111	- every image must earn its place
112	- each image should contribute a distinct job: atmosphere, shape language, color, space, movement, HUD/readability, collectible/reward language, material finish, or palette reference
113	- reject anything that is merely "related"
114	- prefer one strong image per anchor source/title over multiple samey screenshots unless a sequence is the point
115	- when the user wants a cleaner board, bias toward stronger composition and palette images even if they are less explanatory
116
117	Write down at least:
118	- why the strong

People who liked this also viewed

6 picks

Type