semantic-kernel

Semantic Kernel is a model-agnostic SDK from Microsoft designed to help developers build, orchestrate, and deploy AI agents and multi-agent systems. It supports Python 3.10+, .NET 10.0+, and Java JDK 17+, and runs on Windows, macOS, and Linux. Key features include: - **Model Flexibility**: Built-in support for OpenAI, Azure OpenAI, Hugging Face, NVIDIA, and local models via Ollama, LMStudio, or ONNX. - **Agent Framework**: Create modular AI agents with access to tools/plugins, memory, and planning capabilities. - **Multi-Agent Systems**: Orchestrate workflows with collaborating specialist agents. - **Plugin Ecosystem**: Extend agents with native code functions, prompt templates, OpenAPI specs, or Model Context Protocol (MCP). - **Vector DB Support**: Integration with Azure AI Search, Elasticsearch, Chroma, and more. - **Multimodal Support**: Process text, vision, and audio inputs. - **Process Framework**: Model complex business processes with a structured workflow approach. - **Enterprise Ready**: Built for observability, security, and stable APIs. Installation is straightforward: `pip install semantic-kernel` for Python, `dotnet add package Microsoft.SemanticKernel` for .NET, and Java instructions are available in the repository. Quickstart examples demonstrate creating a basic chat agent, adding plugins for custom tools, and building multi-agent systems with triage, billing, and refund agents. Note: The project has evolved into Microsoft Agent Framework (MAF), its enterprise-ready successor, now at version 1.0 with stable APIs and long-term support. A migration guide is available for existing Semantic Kernel users. For further learning, the documentation offers a getting started guide, over 100 detailed samples, and API references for C# and Python. Community support is available via GitHub issues, discussions, and Discord.