About this project

The OpenAI Agents SDK is a framework designed for creating multi-agent workflows. It is provider-agnostic, supporting OpenAI's Responses and Chat Completions APIs as well as over 100 other LLMs. Key capabilities include: - **Agent Types**: Supports standard text agents, Sandbox agents for long-horizon tasks (file inspection, command execution), Realtime agents for low-latency multimodal experiences via WebSockets, and Voice agents for speech-to-text/text-to-speech pipelines. - **Workflow Orchestration**: Features handoffs and 'agents as tools' for delegating tasks between different agents. - **Tooling & Safety**: Integration with various tools (functions, MCP, hosted tools) and configurable guardrails for input/output validation. - **Operational Features**: Built-in session management for conversation history, human-in-the-loop mechanisms, and tracing for debugging and optimizing workflows. The SDK requires Python 3.10+ and offers optional extensions for voice support and Redis-based session management.