Agents can plan and complete tasks using tools, work with other agents, and maintain context across steps. Choose a runtime based on where you want orchestration to run and who should manage the state between tasks.
Choose your starting point
| You want to | Start here |
|---|---|
| Run an agent with the Codex harness managed by OpenAI | Agents API |
| Control the agent loop in your application with reusable agents, tools, and handoffs | Agents SDK |
| Work directly with model responses and control your integration | Responses API |
| Add an embedded chat experience | ChatKit |
Compare agent runtime options
| Agents API | Agents SDK | Responses API | |
|---|---|---|---|
| Use for | Long-running tasks where OpenAI manages the agent and saves its progress | Building agents with custom tools and workflows in your application | Calling models directly or building an agent from scratch |
| Where the agent runs | OpenAI runs a managed Codex harness | The SDK runs inside your application | Your application, with optional hosted orchestration |
| Agent integration effort | Low | Medium | High |
| State between tasks | Saved session configuration, turns, and items | Your storage and SDK sessions, or Responses conversation state | Manual history, response chaining, or Conversations |
| Tool execution | Service-connected tools, application function handlers, and an optional sandbox | Tools and integrations configured in your application | Hosted tools and tools your application runs |
| Execution environment | OpenAI hosted sandbox, self-hosted sandbox, or no sandbox | Your runtime and sandbox provider integrations | Your own execution environment |
| Start here | Agents API overview | Agents SDK overview | Responses guide |
The Agents API runs the Codex harness and manages the underlying agent infrastructure so you can focus on what your agents do. It includes automatic context compaction, multi-agent orchestration, programmatic tool calling, and support for MCP servers. See Architecture.
The Agents SDK gives your application control over deployment, storage, approvals, and runtime integration. Its runner handles the agent loop and handoffs. See Running agents.
Add tools, skills, and prompt caching
Tool design, reusable skills, and prompt caching apply across agent workflows. Their configuration and lifecycle can differ by API.
- Start with Using tools for function calling, MCP, and hosted capabilities.
- Read Programmatic Tool Calling for orchestration with JavaScript and the configuration for each API.
- Use Skills for reusable instructions and the supported loading mechanisms.
- Read Prompt caching for shared caching behavior, then Agents API observability and usage for session accounting.
An Agents API session, an SDK session, a Responses conversation, and a sandbox are different resources. Follow the state and cleanup instructions for the runtime you choose.