MYND - Advanced AI Agent Platform with GUI Automation, Persistent Memory, Multi-Agent Orchestration, and Skills Marketplace
MYND is a cutting-edge AI Agent Platform designed to enable sophisticated autonomous agent operations. It provides a complete ecosystem for building, deploying, and managing AI agents with human-like capabilities.
- 🤖 GUI Agent: Advanced UI automation with computer vision and stealth mode
- 🧠 Memory System: Persistent vector-based memory with semantic search and importance scoring
- 🎭 Multi-Agent Orchestration: Graph-based workflow engine with specialist agent roles
- 🛒 Skills Marketplace: Extensible skill ecosystem with MCP protocol validation
- ⚡ Production Ready: Dockerized deployment, CI/CD, monitoring, and security
┌─────────────────────────────────────────────────────────────────┐
│ MYND Platform │
├─────────────────────────────────────────────────────────────────┤
│ Frontend (SvelteKit 5 + Tailwind CSS) │
│ ┌──────────┬──────────┬──────────────┬──────────────────┐ │
│ │ Dashboard│ GUI Agent│ Memory System│ Skills Marketplace│ │
│ └──────────┴──────────┴──────────────┴──────────────────┘ │
├─────────────────────────────────────────────────────────────────┤
│ Backend (Fastify + TypeScript) │
│ ┌─────────────┬─────────────┬────────────────┬─────────────┐ │
│ │ GUI Agent │ Memory │ Orchestrator │ Marketplace│ │
│ └─────────────┴─────────────┴────────────────┴─────────────┘ │
├─────────────────────────────────────────────────────────────────┤
│ Data Layer │
│ ┌──────────────┬──────────────┬────────────────────────────┐ │
│ │ PostgreSQL │ Redis │ TimescaleDB │ │
│ │ (pgvector) │ (caching) │ (time-series) │ │
│ └──────────────┴──────────────┴────────────────────────────┘ │
└─────────────────────────────────────────────────────────────────┘
- Node.js 20+
- Docker & Docker Compose
- Git
# Clone the repository
git clone https://github.com/yethikrishna/mynd-platform.git
cd mynd-platform
# Install dependencies
npm install
# Start all services with Docker
docker compose up -d
# Or run locally
npm run dev- Frontend: http://localhost:3000
- Backend API: http://localhost:8000
- API Documentation: http://localhost:8000/docs
UI-TARS powered GUI automation with computer vision capabilities.
import { GuiAgent } from '@mynd/gui-agent';
const agent = new GuiAgent({ headless: false });
await agent.initialize();
await agent.navigate('https://example.com');
await agent.executeTask([
{ action: 'click', target: 'Sign In' },
{ action: 'type', target: 'Email', value: 'user@example.com' }
]);Features:
- Stealth browser with anti-detection
- Human-like mouse/keyboard simulation
- Intelligent element detection
- Screenshot analysis with computer vision
Claude-mem inspired persistent memory with vector-based retrieval.
import { MemoryManager } from '@mynd/memory';
const memory = new MemoryManager(DB_URL, OPENAI_KEY, 'agent-001');
await memory.initialize();
// Store memories
await memory.addMemory('User prefers dark mode', 'fact', { tags: ['preferences'] });
// Semantic search
const context = await memory.getContext('What are the user preferences?');Features:
- Vector embeddings with pgvector
- Importance scoring with time-based decay
- Cross-session knowledge consolidation
- Semantic similarity search
LangGraph + CrewAI inspired multi-agent orchestration.
import { Orchestrator, ResearcherAgent, WriterAgent } from '@mynd/orchestrator';
const orchestrator = new Orchestrator();
orchestrator.addAgent(new ResearcherAgent('agent-1'));
orchestrator.addAgent(new WriterAgent('agent-2'));
// Run complex task with auto-decomposition
const result = await orchestrator.runComplexTask(
'Write a comprehensive report on AI trends 2025'
);Features:
- Graph-based workflow engine
- Specialist agent roles (Researcher, Writer, Coder, Analyst)
- Dynamic task decomposition
- Structured inter-agent communication
Agent-skills + clawhub pattern skill ecosystem.
import { Marketplace } from '@mynd/skills-marketplace';
const marketplace = new Marketplace();
// Search for skills
const results = marketplace.search('web-scraping', { category: 'automation' });
// Install a skill
marketplace.install(skillId, { apiKey: 'your-key' });Features:
- Skill publishing and discovery
- MCP (Model Control Protocol) validation
- Rating and review system
- Automatic dependency management
mynd-platform/
├── frontend/ # SvelteKit 5 Frontend
│ ├── src/
│ │ ├── routes/ # Page routes
│ │ └── lib/
│ │ └── components/ # Reusable components
├── backend/ # Fastify Backend
│ └── src/
│ ├── routes/ # API endpoints
│ └── server.ts
├── packages/ # Shared Packages
│ ├── gui-agent/ # @mynd/gui-agent
│ ├── memory/ # @mynd/memory
│ ├── orchestrator/ # @mynd/orchestrator
│ └── skills-marketplace/ # @mynd/skills-marketplace
├── docker/ # Docker configuration
├── docs/ # Documentation
└── .github/workflows/ # CI/CD pipelines
npm run build --workspacesnpm test --workspaces# Build production images
docker compose -f docker-compose.yml build
# Start production services
docker compose -f docker-compose.yml up -d
# View logs
docker compose logs -fCopy .env.example to .env and configure:
DATABASE_URL=postgresql://mynd:mynd@localhost:5432/mynd
REDIS_URL=redis://localhost:6379
OPENAI_API_KEY=your-key-herePlease read CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.
This project is licensed under the MIT License - see the LICENSE file for details.
- Documentation: docs/
- Issues: GitHub Issues
- Discussions: GitHub Discussions
Built with ❤️ by the MYND Team