Sharad Khare Learn
Domain

DesignForge AI

DesignForge turns uploaded UI designs into structured feedback across three analysis modes: 1. **Visual Design** — color, typography, layout, hierarchy, and branding 2. **User Experience** — flows, usability, accessibility, and interaction patterns 3. **Market Analysis** — trends, positioning, and competitor insights via web search Upload your designs, select focus areas, and run one or more analysis types in a single pass.

Python Streamlit
← All projects

DesignForge AI

Multimodal design analysis team by Sharad Khare.

DesignForge AI is a three-agent Streamlit app that analyzes UI screenshots with Gemini vision, UX critique, and market research — optionally comparing against competitor designs.

Built and maintained by Sharad Khare — AI strategist, full-stack developer, and creator of practical agent workflows.

What this project does

DesignForge turns uploaded UI designs into structured feedback across three analysis modes:

1. Visual Design — color, typography, layout, hierarchy, and branding

2. User Experience — flows, usability, accessibility, and interaction patterns

3. Market Analysis — trends, positioning, and competitor insights via web search

Upload your designs, select focus areas, and run one or more analysis types in a single pass.

Why Sharad Khare built DesignForge AI

Design reviews span visual craft, usability, and market fit — rarely handled well by a single prompt. DesignForge demonstrates how specialized Agno agents with Gemini vision and DuckDuckGo research produce richer, more actionable design intelligence than a one-shot LLM call.

How it works

UI Screenshots + Optional Competitor Images
   │
   ▼
Vision Agent → Visual Design Analysis
   │
   ▼
UX Agent → User Experience Analysis
   │
   ▼
Market Agent → Market & Competitor Analysis
   │
   ▼
Streamlit results with markdown sections

Agent roles

| Agent | Focus | Output |

|-------|-------|--------|

| Vision | Visual hierarchy, color, typography | Visual Design Analysis |

| UX | Flows, usability, accessibility | UX Analysis |

| Market | Trends, competitors, positioning | Market Analysis |

Features

  • Three-agent Agno team with Gemini multimodal vision
  • Upload multiple UI designs plus optional competitor screenshots
  • Configurable analysis types and focus areas
  • DuckDuckGo-powered market research agent
  • Modular package layout for extension

Quick start

cd designforge-ai
pip install -r requirements.txt
streamlit run app.py

Required API key

  • Gemini API key — enter in the sidebar when running the app

Example inputs

  • Designs: Mobile app onboarding screens (PNG)
  • Competitors: Two rival app screenshots
  • Analysis: Visual Design + User Experience
  • Focus: Color Scheme, Navigation, Accessibility
  • Context: Fintech app targeting first-time investors

Project structure

designforge-ai/
├── app.py
├── designforge/
│   ├── config.py
│   ├── agents.py
│   └── analysis.py
├── requirements.txt
├── pyproject.toml
└── README.md

Who this is for

  • Product designers reviewing UI iterations quickly
  • Founders validating landing page designs before launch
  • UX researchers comparing competitor screenshots
  • Developers building multimodal agent workflows

Use cases

  • Audit color, typography, and layout consistency
  • Identify usability and accessibility gaps
  • Compare your UI against competitor designs
  • Generate market positioning insights for new products

Author

License

MIT © Sharad Khare