MiniMax-AI/OpenAgentCore

The OpenAI Agents API, open source and self-hosted. Run Codex, Claude Code or MiniMax Code on your own machines with your own models, through the official OpenAI SDK.

Website · Documentation · Quickstart · Contributing

English · 简体中文 · 日本語

• Same API as OpenAI. Set OPENAI_BASE_URL and the official OpenAI SDK, or plain HTTP, works unchanged.
• Your agent, your model. Each Session runs Codex, Claude Code or MiniMax Code natively, as it runs on your machine, with the model provider you configure. See Harness capabilities.
• Compute that sleeps, workspaces that stay. On microsandbox with independent workspace storage, idle sandboxes release compute and resume on a compatible node with their files intact. E2B, Docker and your own Linux, macOS or Windows machine also run Sessions. See Sandbox Provider protocol.
• Replaceable parts. Sandbox Providers, workspace filesystems, Harnesses and model providers plug in through defined protocols.

Install Core, Web and PostgreSQL on Linux or macOS with Docker:

curl -fsSL https://openagentcore.dev/install.sh | bash

On Windows, in PowerShell:

irm https://github.com/MiniMax-AI/OpenAgentCore/releases/latest/download/install.ps1 | iex

In Web, sign in with the Core key the installer prints, configure the public address, set a default model, issue a Project API key and add sandbox capacity (a node or E2B). Install Core and Web covers each step.

Then call it with the OpenAI SDK, with OPENAI_BASE_URL set to your API base URL (https://core.example/v1) and OPENAI_API_KEY to the Project API key:

from openai import OpenAI

client = OpenAI()  # reads OPENAI_BASE_URL and OPENAI_API_KEY
session = client.beta.agents.sessions.create(
    environment={"type": "openai_hosted"},
    input="Create /workspace/hello.txt with a short greeting, then describe it.",
    extra_body={"agent": {"x_agents_core": {"harness": "codex"}}},
)
print(session.id)

Run your first Session waits for the result and reads it back.

Applications and operators use these Core APIs:

Core keeps durable execution state. The Runtime runs the chosen Harness inside the Environment. Each connection is a defined protocol, so any part can be replaced on its own. See Architecture.

Read CONTRIBUTING.md before changing code. Report bugs and ask questions in GitHub Issues. Released under the MIT License.