Your AI-Powered Kanban Board for Task Management with AI Agents
Where AI meets productivity in a familiar Kanban interface
AgentsBoard is a JIRA/Trello-inspired task management system designed for AI agents. Easily input ideas, assign them to AI agents, and track their execution across Todo, In Progress, and Done states.
- 📝 Effortless Task Creation: Simple interface for task creation and agent assignment
- 🤖 AI Agent Integration: Powered by OpenRouter for access to top AI models
- 🔄 Automated Task Lifecycle: Tasks automatically progress through states
- 📊 Intuitive Kanban Board: Clear visualization of task progress
- 💾 Markdown Export: Download completed tasks in markdown format
- 🎨 Modern UI: Built with Next.js 14, Tailwind CSS, and shadcn/ui
Try AgentBoard now: https://agentsboard.vercel.app
Deploy your own instance of AgentBoard with one click:
- Clone the repository:
git clone https://github.com/Justmalhar/agent-board.git
cd agent-board
- Install dependencies:
npm install
- Set up environment variables:
cp .env.example .env.local
- Add your OpenRouter API key to
.env.local
:
NEXT_PUBLIC_OPENROUTER_API_KEY=your_api_key_here
NEXT_PUBLIC_APP_NAME=AgentBoard
NEXT_PUBLIC_DEFAULT_MODEL=liquid/lfm-40b:free
OPENROUTER_BASE_URL=https://openrouter.ai/api/v1
MAX_REQUESTS_PER_MINUTE=60
- Start the development server:
npm run dev
AgentBoard/
├── app/
│ ├── api/ # API routes for task execution
│ ├── layout.tsx # Root layout
│ └── page.tsx # Home page
├── components/ # React components
│ ├── KanbanBoard.tsx
│ ├── TaskCard.tsx
│ └── TaskInput.tsx
├── contexts/ # React contexts
│ └── TaskContext.tsx
├── lib/ # Utility functions
│ └── agents.ts
├── prompts/ # Predefined AI prompts
│ ├── default.md
│ └── ai.md
└── public/ # Static assets
AgentBoard comes with predefined prompts in the /prompts
directory thanks to Fabric Patterns and Awesome ChatGPT-Prompts
AgentBoard uses OpenRouter as the AI provider, giving you access to:
- Multiple AI models (Claude, GPT-4, Llama, etc.)
- Flexible token limits
- Cost-effective API calls
-
Create a Task:
- Enter task description
- Select an AI agent
- Choose model and token length
- Submit task
-
Task Execution:
- Task appears in Todo column
- Click Play to execute
- AI processes the task
- Results appear in Done column
-
View Results:
- Preview results in markdown
- Download as .md file
- Track task status
Contributions are welcome! Please feel free to submit a Pull Request.
This project is licensed under the MIT License - see the LICENSE file for details.
Built with ❤️ and AI by Malhar Ujawane
Follow me: