AI Agent Platforms / All platforms / OpenAI Agents SDK
OpenAI Agents SDK
Python framework for multi-agent workflows with tools, handoffs, guardrails, and tracing.
The OpenAI Agents SDK is a Python framework for building agent and multi-agent workflows. It is provider agnostic, working with OpenAI APIs and 100+ other LLMs. Developers use it to compose agents with instructions, tools, guardrails, and handoffs, including voice and realtime agents.
| Category | Agent builders and frameworks |
| License | MIT |
| Hosting | library for self-managed applications |
| Repository | github.com/openai/openai-agents-python |
| Website | openai.github.io/openai-agents-python |
| Last verified | 2026-09-02 |
Documented features
- Configures agents with instructions, tools, guardrails, and handoffs
- Hands off control between agents and can use agents as tools for other agents
- Integrates function tools, MCP servers, and hosted tools
- Guardrails validate agent inputs and outputs
- Human-in-the-loop mechanisms during agent runs
- Automatic session management for conversation history
- Built-in tracing for viewing and debugging workflows
- Sandbox agents can inspect files, run commands, and preserve workspace state
- Realtime and voice agents combine speech-to-text, workflows, and text-to-speech
Feature summaries come from the project's own repository and documentation and describe the software as documented on the verification date, not an evaluation.