Brand Kit Generator

A research-driven skill that creates complete brand identities — logo, color palette, typography, imagery, domain names, and social media handles — with real-time availability checking. Goes beyond generic branding by researching the industry, competitors, and audience before generating any assets.

This is NOT a logo generator. It's a full brand identity system: research the space, define the strategy, generate visual assets, verify digital availability, and deliver a polished brand kit.

Workflow Overview

Input (business idea, existing materials)
  -> Research (industry, competitors, audience)
    -> Brand Strategy (positioning, voice, personality)
      -> Visual Identity (colors, typography, mood)
        -> Logo Generation (AI-generated options)
          -> Brand Imagery (hero, social, patterns)
            -> Domain Check (suggest + verify availability)
              -> Username Check (suggest + verify across platforms)
                -> Brand Kit Delivery (Notion / Drive / PDF)

Step 0: Gather Inputs

Check for existing materials first:

"Do you have any existing materials I should start from? This could be a brief, mood board, existing logo, website, social profiles, or any document describing the brand. Share the file path, paste the content, or provide URLs — otherwise I'll walk you through it step by step."

If materials are provided:

• Read files using the Read tool
• Scrape URLs using

  FirecrawlScrapeTool

• Fetch social profiles using

  FetchInstagramProfileTool

  ,

  FetchTiktokProfileTool

  ,

  FacebookBusinessPageInfoTool

• Extract: brand name, industry, existing colors, tone, audience, visual style
• Present a summary table and confirm before proceeding. Skip to Step 2.

If no materials — ask these in a single message:

Step 1: Research Phase

Run all applicable research in parallel.

1A: Industry & Competitor Research

Social Toolkit MCP:
- PerplexitySonarSearchTool -> "[industry] branding trends 2025" (current visual trends)
- PerplexitySonarSearchTool -> "best [industry] brands [region]" (competitor landscape)
- GoogleNewsSearchTool -> "[industry] brand design" (recent trends, rebrands)

1B: Competitor Visual Analysis

For 2-3 top competitors identified in 1A:

Social Toolkit MCP:
- FirecrawlScrapeTool -> competitor websites (extract color schemes, typography, tone)
- FetchInstagramProfileTool -> competitor handles (visual style, content patterns)

Extract from each competitor:

• Color palette (dominant, accent, neutral)
• Typography style (serif, sans-serif, display)
• Visual tone (photography vs illustration, dark vs light, busy vs minimal)
• Logo style (wordmark, symbol, combination, emblem)
• Brand voice (formal, playful, technical, warm)

1C: Audience Research

Social Toolkit MCP:
- GoogleForumsSearchTool -> "[target audience] preferences [industry]" (what resonates)
- PerplexitySonarSearchTool -> "[target audience] demographics design preferences"

Compile a research summary:

"Here's what I found about the [industry] landscape: [key findings]. The main competitors use [patterns]. The target audience responds to [preferences]. I'll use this to differentiate the brand. Ready to proceed?"

Step 2: Brand Strategy

Based on research, define the brand strategy. Present as a structured summary:

Brand Positioning

Visual Direction

Based on personality and audience, propose a visual direction:

Present all three options with a brief mood description. Ask the user to pick one or combine elements.

"Which direction feels right? Pick one, or tell me what to combine from each."

Step 3: Visual Identity

3A: Color Palette

Generate a complete palette based on the chosen direction:

#XXXXXX

#XXXXXX

#XXXXXX

#XXXXXX

#XXXXXX

#XXXXXX

Include contrast ratios for accessibility (WCAG AA minimum).

3B: Typography

Recommend a font pairing (Google Fonts for accessibility):

3C: Visual Style Guide

Define the overall aesthetic rules:

• Photography style (if applicable): [e.g., natural light, high contrast, desaturated]
• Illustration style (if applicable): [e.g., flat, isometric, hand-drawn, geometric]
• Icon style: [e.g., outlined, filled, duotone]
• Corner radius: [e.g., sharp, slightly rounded (4px), fully rounded (12px+)]
• Spacing philosophy: [e.g., generous whitespace, compact and dense]
• Pattern/texture: [e.g., subtle grain, geometric patterns, clean flat]

Step 4: Logo Generation

Generate logo options using available image generation tools.

Check for

infsh

which infsh

If available, generate logos using the model best suited for graphic design:

infsh app run falai/flux-dev-lora --input '{
  "prompt": "[logo prompt based on brand strategy]"
}'

If

infsh

HiggsfieldImageTool

nano_banana_pro

HiggsfieldImageTool:
  model: "nano_banana_pro"
  prompt: "[logo prompt]"
  aspect_ratio: "1:1"
  resolution: "1080p"

Generate 3-4 logo variants:

Prompt engineering for logos:

• Include the brand personality traits in every prompt
• Specify "logo design, vector style, clean lines, [brand colors], transparent background"
• Reference the visual direction chosen in Step 2
• Avoid photorealism — logos need to work as flat graphics
• Specify "no text" for symbol variants to avoid garbled AI text

Download all generated logos to the output directory.

Step 5: Brand Imagery

Generate supporting brand images for use across web and social.

Image set to generate:

nano_banana_2

infsh

nano_banana_pro

nano_banana_2

infsh

nano_banana_2

infsh

soul_2

infsh

Prompt engineering for brand images:

• Use the brand's color palette explicitly in prompts
• Match the visual style guide from Step 3C
• Reference the industry and audience context from research
• Never generate generic stock-photo-style images

Fallback chain:

1. infsh

   CLI (preferred — FLUX or Grok models)

2. HiggsfieldImageTool

   (Social Toolkit MCP)
3. CSS-only effects + detailed prompts saved for manual generation later

Download all generated images to the output directory alongside the logos.

Step 6: Domain Name Suggestions

Suggest 8-12 domain name options based on the brand name, positioning, and industry.

Naming strategies to cover:

brandname.com

get[brand].com

try[brand].com

hello[brand].com

[brand]app.com

[brand]hq.com

[brand]studio.com

brandname.io

brandname.co

brandname.dev

brandname.shop

Check availability for each domain by fetching the URL and checking for non-200 responses:

WebFetch -> https://[domain]

A connection error, 404, or "domain not found" page indicates the domain is likely available. A 200 with real content indicates it's taken. A parked/for-sale page means it's registered but potentially purchasable.

Present results as a table:

brandname.com

brandname.io

getbrandname.com

brandname.co

Highlight the top 2-3 recommendations with reasoning.

Step 7: Social Media Username Check

Suggest 5-8 username options and check availability across platforms.

Platforms to check:

https://www.instagram.com/[username]/

https://www.tiktok.com/@[username]

https://x.com/[username]

https://www.youtube.com/@[username]

https://www.facebook.com/[username]

https://github.com/[username]

https://www.linkedin.com/company/[username]

Check availability by fetching each URL:

WebFetch -> [platform URL with username]

A 404 response or "page not found" content indicates the username is likely available. A 200 with a real profile means it's taken.

Run all platform checks for each username in parallel to minimize wait time.

Present results as a matrix:

brandname

brandname_official

getbrandname

Highlight usernames that are available across ALL platforms. Recommend the most consistent handle.

"My top recommendation is

@[username]

— it's available on all [N] platforms and matches the domain

[domain]

. Want me to finalize the brand kit with these?"

Step 8: Brand Kit Delivery

Compile all brand assets into a deliverable. Ask the user which output format to use:

"How would you like the brand kit delivered?"

1. Notion — creates a structured Notion page with all assets, colors, fonts, and availability results
2. Google Drive — generates an HTML brand guide + asset folder ready to upload
3. Download (PDF-ready) — generates a self-contained HTML brand guide, opens in browser for PDF export via print

Option 1: Notion

Use

notion-create-pages

• Brand Strategy (positioning, personality, voice)
• Color Palette (with hex swatches)
• Typography (font names, weights, pairings)
• Logo Options (embedded images)
• Brand Imagery (embedded images)
• Domain Recommendations (availability table)
• Social Media Handles (availability matrix)
• Visual Style Guide (rules and guidelines)

Option 2: Google Drive

Generate a folder structure:

brand-kit-[brandname]/
+-- brand-guide.html          # Visual brand guide (self-contained)
+-- logos/
|   +-- logo-symbol.png
|   +-- logo-wordmark.png
|   +-- logo-combination.png
|   +-- logo-monogram.png
+-- images/
|   +-- hero-banner.png
|   +-- social-avatar.png
|   +-- og-share.png
|   +-- pattern-texture.png
|   +-- lifestyle-1.png
|   +-- lifestyle-2.png
+-- availability-report.md    # Domain + username availability results

Save to

/tmp/brand-kit-[brandname]/

Option 3: Download (PDF-ready)

Generate a single self-contained HTML file with:

• All images embedded as base64 or referenced from downloaded files
• Print-optimized CSS (

  @media print

  with page breaks)
• Clean layout: cover page, strategy, palette swatches, typography samples, logo showcase, imagery grid, availability tables
• Open in the default browser with instructions: "Print to PDF (Cmd+P / Ctrl+P) to save"

Save to

/tmp/brand-kit-[brandname]/brand-guide.html

Conventions

• Research before creating — never generate logos or suggest names without understanding the industry and audience first.
• Parallel operations — run all social media checks, image generations, and domain lookups in parallel where possible.
• Confirm at gates — confirm brand strategy (Step 2) and visual direction before generating assets.
• Real availability data — always check domains and usernames live. Never guess availability.
• Download all assets — every generated image must be downloaded to the output directory, not just linked.
• Language detection — default to auto-detecting language from the user's input. Support explicit override.