BrainBlend-AI/atomic-agents
A modular Python framework for composing AI agents from atomic, reusable components like agents, tools, and context providers.

Atomic Agents is a framework designed around the concept of atomicity for building agentic AI pipelines and applications. It provides modular, composable components that serve as building blocks for constructing AI applications—each component (agent, tool, context provider) can be used independently or combined with others. The framework emphasizes developer experience and maintainability while supporting integration with LLM providers like OpenAI.