Cognotik: A New FOSS AI Coding Assistant. For JetBrains IDEs

Cognotik is a comprehensive AI-powered development platform that combines intelligent planning, code generation, and interactive interfaces to streamline software development workflows. The platform consists of multiple integrated components that work together to provide a complete solution for AI-assisted development.

For detailed installation instructions, see the Installation Guide.

Cognotik is 100% open source software, released under the Apache 2.0 license. The platform follows a "Bring Your Own Key" (BYOK) model, meaning you provide your own API keys for AI services (OpenAI, Anthropic, etc.). This gives you complete control over:

• Which AI providers and models you use
• Your data privacy and security
• Your usage costs and billing
• Service configuration and customization

Cognotik uses a "Bring Your Own Key" (BYOK) model for all AI service integrations:

• You must provide your own API keys for services like OpenAI, Anthropic, Google AI, etc.
• All API usage costs are billed directly to your accounts with these providers
• No data is shared with any third parties without your explicit configuration
• The software includes usage tracking tools to help you monitor your API consumption

• OpenAI (GPT-4, GPT-3.5, DALL-E)
• Anthropic (Claude 3, Claude 2)
• Google (Gemini, PaLM)
• AWS Bedrock (Various models)
• Azure OpenAI (GPT models)
• Groq (Fast inference)
• Mistral AI (Mistral models)
• DeepSeek (Coding models)
• Perplexity (Search-optimized models)
• Local Models (Ollama, LM Studio)

The foundation of the platform, providing essential services and utilities for AI interactions.

Key Features:

• Actor system for AI model interactions (SimpleActor, CodingActor, ImageActor)
• Session management and data persistence
• Authentication and authorization
• Code patch generation and application utilities
• Extensible interpreter framework for code execution
• Token counting and cost estimation
• Rate limiting and retry logic

A framework for building interactive web applications with real-time communication.

Key Features:

• Chat system with real-time communication
• Session management and state persistence
• Interactive UI components (links, inputs, tasks, file handling)
• Application framework for building web interfaces
• Local storage management
• File upload and download capabilities
• Session history and management

AI-assisted task planning and execution framework.

Key Features:

• Task types for file modifications, code generation, and web searches
• Plan coordination with dependency management
• Plan visualization tools
• Multiple cognitive modes (AutoPlanMode, PlanAheadMode, GoalOrientedMode, TaskChatMode, GraphOrderedPlanMode)
• Self-healing task execution

A standalone desktop application that hosts the Cognotik platform.

Key Features:

• System tray integration for easy access
• Background daemon process
• Socket-based communication for remote control
• Cross-platform support (Windows, macOS, Linux)
• Auto-update functionality

A React-based chat application interface with real-time messaging.

Key Features:

• WebSocket connectivity for real-time updates
• Multiple themes with dynamic switching
• Markdown support with syntax highlighting
• Tab system with state persistence
• Event-driven architecture for real-time updates

An IntelliJ-based plugin that integrates Cognotik capabilities into the IDE.

Key Features:

• Smart code operations (paste, transformations)
• Contextual AI chat with code understanding
• Intelligent workflows for multi-step changes
• Test result autofix and problem analysis
• Code review and documentation generation
• Refactoring suggestions

A unified, type-safe model registry and API for working with AI models.

Key Features:

• Type-safe referencing of models by name or provider
• Centralized pricing and quota logic
• Support for multiple providers (OpenAI, Anthropic, Google, AWS, etc.)
• Unified API for text, chat, embedding, image, and audio models

The Cognotik platform follows a modular architecture:

1. Core Services Layer: Provided by the Core module, handling AI interactions, session management, and data persistence.

2. Application Layer: Consists of the Web UI Framework and Planning Framework, providing the building blocks for AI-powered applications.

3. Client Applications: Desktop Application, Web Application, and IntelliJ Plugin, offering different interfaces to access the platform.

4. Model Integration Layer: JOpenAI provides a unified interface to various AI models and providers.

Communication between components:

• RESTful APIs and WebSockets for client-server communication
• Socket-based communication for desktop application components
• File-based storage for session data and generated content

• Java 17 or higher
• Node.js 16+ (for web application)
• Gradle 7.6+ (for building)
• API keys for supported AI providers (OpenAI, Anthropic, etc.)

1. Download the appropriate package for your platform:

   • Windows: Download MSI installer
   • macOS: Download DMG package
   • Linux: Download DEB/RPM package

2. Run the installer and follow the instructions.

3. Launch the application from your desktop or start menu.

4. Configure your API keys in Settings → API Configuration.

1. Clone the repository

2. Install dependencies:

3. Configure environment variables:

   cp .env.example .env # Edit .env with your API keys

4. Start the development server:

5. Open your browser and navigate to http://localhost:3000

1. Install from the JetBrains Marketplace:
   • Open IntelliJ IDEA
   • Go to Settings/Preferences > Plugins
   • Search for "Cognotik"
   • Click Install and restart the IDE
2. Configure API keys:
   • Go to Settings/Preferences > Tools > Cognotik
   • Enter your API keys for the providers you want to use

1. Clone the repository

2. Build using Gradle:

3. For specific components:

   # Desktop application ./gradlew :desktop:build # IntelliJ plugin ./gradlew :intellij:buildPlugin # Web application cd webapp npm run build

The platform provides several extension points for customization:

1. Actors: Create custom actors for specialized AI interactions
2. Interpreters: Add support for new programming languages
3. Storage Providers: Implement custom storage solutions
4. Authentication Providers: Integrate with identity providers
5. UI Components: Create custom UI components for the web interface

Use the IntelliJ plugin or web interface to:

• Generate code from natural language descriptions
• Refactor existing code with AI assistance
• Get explanations and documentation for complex code
• Fix bugs and implement features with AI guidance

Use the planning framework to:

• Break down complex development tasks into manageable steps
• Generate implementation plans with proper dependencies
• Execute tasks with AI assistance
• Visualize and track progress

Use the platform's knowledge tools to:

• Extract structured data from documentation
• Create searchable knowledge indexes
• Generate documentation from code
• Visualize document relationships

Leverage AI for:

• Security vulnerability detection
• Performance optimization suggestions
• Code quality metrics
• Architecture analysis
• Dependency management

• 📚 Documentation: Full documentation
• 💡 Examples: Sample projects and use cases
• **💬 Community **: Discord Server | GitHub Discussions
• 🐛 Issue Tracking: GitHub Issues
• 📧 Contact: Email Support

We welcome contributions! Please see our Contributing Guide for details on:

• Code of Conduct
• Development setup
• Submitting pull requests
• Reporting issues
• Feature requests

This project is licensed under the Apache 2.0 License - see the LICENSE file for details. As open source software, you are free to use, modify, and distribute the code according to the terms of the license.

• OpenAI, Anthropic, and other AI providers for their powerful models
• JetBrains for the IntelliJ platform
• The open-source community for various libraries and tools used in the project
• All contributors who have helped improve Cognotik

See our public roadmap for upcoming features and improvements.

Made with ❤️ by the Cognotik Team