Features
Local autonomous coding agent that reads, writes, and fixes code
Multi-provider support: OpenAI, Claude, Gemini, Ollama, DeepSeek, Groq, Mistral, etc.
Block-based prompt editor for creating and managing prompt templates
Agentic flavours – custom agent configurations
Task tracker with step-by-step logging
Expert Mode for smarter model review and improved results
Visual diffs for code changes
Multi-project support with tabs
Bring Your Own Keys – no usage fees, only pay the model provider
Unsafe-mode flag for unrestricted file execution
Multiple themes: Light, Caramel Latte, Dark Coffee
Qwen3 model support via MLX on Mac
Three-slot model pipeline for combining models
Role-based agent assignment (PM, architect, engineer, QA)
Structured output generation (requirements, design, code)
Data Interpreter for data analysis tasks
SELA module for self-evolving agents
Multi-agent collaboration and workflow orchestration
Meta-programming support for custom agent behaviors
Modular and extensible architecture
Built-in demo projects and case studies
MIT License fully open-source
Flow orchestration for complex agent pipelines
Integration with GitHub Actions and CI/CD
Community-driven with GitHub and Discord
