Branding Agent

Enter a brand URL to extract colors, typography, and page copy with Firecrawl. A grounded ToolLoopAgent then checks contrast, generates palette variations, and describes writing voice from the scraped page.

Use this when you want a grounded agent that pulls a site's brand tokens and answers contrast, palette, and voice questions from the scraped page.

Loading preview…

Problems solved

Technical challenges this implementation handles out of the box.

Extract brand colours and typography from a website with Firecrawl
Check WCAG contrast for brand colour pairs
Generate palette variations from extracted brand colours
Describe writing voice from scraped page copy with quotes

Use cases

Products and workflows this pattern is designed to support.

Brand design system extraction
Accessibility-minded palette exploration
Contrast checks against live brand colours
Writing-voice audits from a brand homepage

Compare with

Similar patterns, and when each one is the better fit.

Setup

Requirements, wiring steps, and what this pattern adds to your project.

Getting started

Pick how you want to pull this pattern in. Then wire env vars and routes the same way.

1
Download → Next.js app
Use Download → Next.js app in the preview toolbar. You get a zip scaffold (Pro downloads rewrite preview `/view/.../api/*` paths to `/api/*` automatically).
2
Extract the zip and install dependencies
Unzip the archive, then run `pnpm install` (or npm / yarn) in the project folder.
3
Fill .env.local
The download includes .env.local placeholders when needed. Add real values for: OPENAI_API_KEY, FIRECRAWL_API_KEY.
4
Customize the agent and tool files
Adapt prompts, tools, and stop conditions in the generated project for your product.
5
Run the app and open the pattern route
Run `pnpm dev`, open the setup/checklist page if present, then navigate to the generated pattern route and verify responses.

Environment variables

OPENAI_API_KEYGet key
FIRECRAWL_API_KEYGet key

Capabilities

AI SDK APIs

useChatgatewayToolLoopAgenttoolhasToolCallisStepCounttoolsContextcreateAgentUIStreamResponseInferAgentUIMessageDefaultChatTransportUIMessage

Providers

OpenAI

External services

FirecrawlUpstash

Files added

20 files

app/page.tsx
app/layout.tsx
app/api/branding-agent/route.ts
lib/actions.ts
lib/brand-context.ts
lib/branding-types.ts
lib/color.ts
lib/contrast.ts
lib/grounding.ts
lib/labels.ts
lib/normalize-brand.ts
lib/rate-limit.ts
lib/scrape-brand.ts
lib/tokens.ts
lib/ai/branding-agent.ts
lib/ai/resolve-color.ts
lib/ai/tools/check-contrast.ts
lib/ai/tools/describe-voice.ts
lib/ai/tools/generate-palette.ts
README.md

Dependencies

10 total

npm packages

AISDK@mendable/firecrawl-jsUpstashlucide-reactNext.jsReactsonnerZod

Registry components

buttoninput

Critical files

7 files

branding-agent/route.ts
API route that builds the agent per request and streams via createAgentUIStreamResponse with brand + pageText.
ai/branding-agent.ts
ToolLoopAgent factory with 3 grounded tools and toolsContext injection from scraped brand + page copy.
lib/scrape-brand.ts
Firecrawl scrape pipeline that normalizes brand tokens and page text for the agent.
tools/check-contrast.ts
WCAG contrast check tool that resolves role names or hex against brand context.
tools/generate-palette.ts
Palette tool for shade scales, complementary, and analogous colours from a base brand colour.
tools/describe-voice.ts
Voice tool that describes page copy tone with quoted evidence from scraped text.
lib/actions.ts
Server action for initial brand extraction from a URL before the agent chat starts.

Requirements

Complexity: Advanced
Setup time: ~10 minutes
Node: 24+
Memory: 256MB

Upstash Redis

Loading preview…