🤖 Harness SDK: Model-Driven AI Agent Framework

This project provides a highly streamlined approach to building AI agents. Its core philosophy is "model-driven," attempting to let developers transform a base large language model into an autonomous agent with just a few lines of code. By integrating tools like LiteLLM under the hood, it natively supports the vast majority of mainstream models available today. This is not an attempt to build a massive, complex workflow engine; rather, it focuses on lowering the barrier to entry for agent development. For developers who do not need a complex visual interface but simply want to quickly introduce agentic capabilities into an existing Python codebase—allowing models to autonomously call tools and complete specific tasks—this lightweight SDK offers a very convenient option.