Fus3n/TwoAI

Orchestrates multi-agent conversations using Ollama as a local inference backend, allowing independent model selection and remote host configuration per agent. The framework manages turn-based exchanges with configurable exit conditions and context windows, enabling use cases from debates to collaborative code review. Built as a Python library with minimal dependencies, it abstracts the complexity of prompt engineering and conversation state management for coordinated LLM interactions.

113 stars. No commits in the last 6 months.