Ai campaign studio
Platform-agnostic AI skill: transform a brand brief into a complete visual campaign — concept, brand system, style anchor, and production-ready marketing assets.From the repository description
npx -y skills add torykovdya/ai-campaign-studioAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
5.9 KB, ~1.3k tokens by cl100k_base, as published. Nobody here has run it
description: Transform a brand or product brief into a complete visual campaign — concept, brand system, style anchor, and production-ready marketing assets. argument-hint: <brand brief: description, audience, tone, campaign goal> allowed-tools: Read, Write, Bash, WebFetch
AI Campaign Studio
Transform a brand or product brief into a complete visual identity and launch asset system. Generate a consistent visual language and production-ready assets for social media, websites, presentations, and advertising.
Use When
- Launching a new brand or product
- Needing a visual campaign concept fast
- Exploring creative directions before production
- Creating visual assets for marketing or pitch
Do Not Use When
- User needs a single illustration or icon
- Brief has no product, brand, or audience defined
- User asks for realistic photography of real people
Inputs
- Brand or product description
- Target audience
- Campaign goal (awareness, launch, emotion, conversion)
- Tone (bold, minimal, playful, premium, rebellious, warm)
- Optional: color preferences, visual references
Rules
- Always define Creative Concept before generating any images
- Color palette must be established in Step 2 — never change it after
- All campaign assets must share the same visual identity, color palette, lighting style, and brand personality
- Each image must serve a different narrative purpose
- Use the highest-quality available image model. Preserve the same style anchor and generation requirements regardless of model choice.
- Prompts must include: mood, lighting, composition, color palette, atmosphere
- Style Consistency Check is mandatory — never skip it
- Define a Style Anchor in Step 2 — copy it verbatim into every image prompt
- At least 70% of generated assets must visualize actual services or outcomes. No more than 30% may use abstract AI symbolism
- Do not change artistic medium between images (e.g., illustration → photography is prohibited)
Workflow
Step 1 — Brand Analysis
Extract from brief:
- Brand personality (3 adjectives)
- Target audience (who they are, what they value)
- Campaign goal (what should the audience feel or do)
- Emotional tone (how the campaign should feel)
- Key visual metaphors (what imagery represents this brand)
Step 1.5 — Differentiation Strategy
Analyze:
- Common visual clichés in the category
- Overused color palettes
- Overused imagery and metaphors
Output:
- What competitors typically do
- What this campaign should avoid
- Unique visual territory to own
Step 1.6 — Business Visualization Strategy
Translate services into visuals. For each core service define:
- Visual representation
- Metaphor
- Recurring symbol
Examples:
- Social Media → content streams
- Automation → synchronized gears
- Research → intelligence radar
- Video → media frames
Step 2 — Creative Direction
Define the campaign's visual language:
- Campaign concept (1–2 sentences, the big idea)
- Campaign narrative (the story the images tell together)
- Color palette (4–5 hex codes with names)
- Lighting style (soft/dramatic/natural/studio/cinematic)
- Composition style (minimal/dynamic/macro/wide/abstract)
Style Anchor
Output a reusable style block containing:
- Artistic style
- Rendering style
- Color palette
- Lighting style
- Aspect ratio
- Quality keywords
This exact block must be copied verbatim into every image prompt.
Step 2.5 — Brand System
Define:
- Visual Identity summary
- Typography Direction
- Iconography Style
- Brand Voice
Step 3 — Asset Planning
Generate a complete launch asset set.
Required assets:
- Profile Avatar (Founder or Brand Symbol — explain choice)
- Banner / Cover Image
- Website Hero Image
- Social Media Post #1 — Announcement
- Social Media Post #2 — Product/Service highlight
- Social Media Post #3 — Customer outcome
For each asset define:
- Marketing Purpose
- Target Platform
- Dimensions
- Visual Objective
- Scene description
- Emotion triggered
- Camera angle and composition
Visual Continuity: at least one recurring visual element must appear across all assets (same character, product, symbol, or environment).
Asset Mapping: map every asset to a specific business use case.
CRITICAL: Profile Avatar and Banner are mandatory. Do not replace them with other assets.
Step 4 — Asset Generation
For every asset output:
- Asset Name
- Purpose
- Target Platform
- Dimensions
- Full prompt (including verbatim Style Anchor)
Generate all assets using the highest-quality available image model.
Step 5 — Style Consistency Audit
After all assets are generated, verify:
For each category provide Status (PASS / FAIL) and Score (0–100):
- Color Palette
- Lighting Style
- Artistic Style
- Visual Continuity
- Aspect Ratio
- Prompt Consistency
Provide an Overall Consistency Score.
If any category FAILS: flag it and suggest specific prompt adjustment.
Validation
Before final output verify:
- Creative concept defined before any generation
- Color palette established in Step 2 and used in all prompts
- Style Anchor defined and copied verbatim into all prompts
- All required assets generated (Avatar, Banner, Hero, 3 Social Posts)
- Each asset serves a distinct marketing purpose
- Asset Mapping completed
- Social posts communicate different messages (Announcement / Product / Outcome)
- Style Consistency Audit completed with scores
Output
A complete launch asset package:
- Campaign Concept (the big idea)
- Campaign Narrative (the story)
- Differentiation Strategy (what to avoid and own)
- Color Palette (hex codes with names)
- Style Anchor (reusable prompt block)
- Brand System summary
- Asset Mapping (every asset → business objective)
- 6 Generated Assets with individual prompts
- Style Consistency Report with overall score
What ships with it: 8 files
24.3 KB alongside SKILL.md
evals/
- evaluation_criteria.md2.7 KB
- iteration_notes.md3.2 KB
examples/
- architecture.md3.8 KB
- description.md171 B
- edge_cases.md2.8 KB
- LICENSE1.0 KB
- README.md4.1 KB
Gives 0 of the 12 instructions most marketing audience skills give in ~1.3k tokens
Counted across 690 of the 894 authors here whose files we hold, read 2026-08-07
- Apply Poppins font to headingsin 41 of 690, across 6 files
- Apply Lora font to body textin 41 of 690, across 6 files
- Use Arial fallback for headingsin 39 of 690, across 4 files
- Use Georgia fallback for body textin 39 of 690, across 4 files
- Maintain text hierarchy and formattingin 39 of 690, across 4 files
- Use accent colors for non-text shapesin 38 of 690, across 3 files
- Use RGB values for precise color matchingin 38 of 690, across 3 files
- Use brand colors for primary text and backgroundsin 36 of 690, across 1 file
- Read product marketing context file before asking questions, starting, or auditingin 35 of 690, across 23 files
- Use active voice instead of passive voicein 26 of 690, across 10 files
- Implement or generate appropriate JSON-LD structured datain 24 of 690, across 17 files
- Prioritize clarity over clevernessin 22 of 690, across 8 files
Said here and by no other author read
- define creative concept before generating images
- establish color palette before generating assets
- maintain consistent visual identity across assets
- make each image serve a different narrative purpose
- use the highest-quality available image model
- include specific visual elements in each prompt
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.