---
id: daytona
version: "0.204.0"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# daytona — Python SDK for Daytona

License: permissive · Maintenance: active · Downloads: 8.9M/mo

## What it is and what it does

Daytona is the official Python SDK for Daytona, a managed infrastructure service that provisions isolated sandboxed environments for running AI-generated or untrusted code. The SDK exposes a programmatic interface to create, configure, and manage these sandboxes, along with operations for file I/O, Git integration, process execution, and language server protocol support. It includes both synchronous and asynchronous clients, with 24 runtime dependencies covering HTTP clients (aiohttp, httpx), async utilities (aiofiles, python-socketio), observability (OpenTelemetry), and configuration management (pydantic, python-dotenv).

The package is designed for developers building AI agents, code execution platforms, or applications that need to safely run untrusted code in isolated environments. You authenticate with an API key, instantiate a Daytona client, create sandboxes with optional resource constraints, and then interact with them via methods for code execution, file operations, Git operations, and LSP support. The SDK handles the complexity of remote sandbox lifecycle management and provides both paginated and cursor-based list operations for querying resources.

Use it for:

- Build AI agent systems that execute generated code safely in isolated sandboxes without risking the host system.
- Create code execution platforms or playgrounds where users can run untrusted scripts with resource limits and automatic cleanup.
- Develop integration workflows that clone Git repositories, modify files, and execute code within managed sandbox environments.
- Implement language server protocol support for remote code editing and completion in sandboxed development environments.
- Automate testing and validation of generated code by running it in ephemeral, auto-deleted sandbox instances.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Daytona is a Python SDK for managing and interacting with sandboxed computing environments—isolated, on-demand virtual computers that can execute code, run processes, and perform file and Git operations securely.

Yes, with conditions. The package is actively maintained and has low install friction, but it is still in alpha (Development Status :: 3 - Alpha) and requires a Daytona API key and external service access. Install it if you need to programmatically manage sandboxed code execution environments and are comfortable with an alpha-stage SDK. Not suitable if you need a stable, production-grade API or if you cannot rely on external infrastructure.

## Install

pip install daytona
uv add daytona
poetry add daytona

## Installing daytona

Before you install: Low install friction with a pure-Python wheel distribution. Actively maintained with a release 3 days ago. Supports Python 3.10 through 3.14, though the package is still in alpha (Development Status :: 3 - Alpha).

License in practice: Licensed under Apache-2.0 (permissive), allowing commercial use, modification, and distribution with minimal restrictions—suitable for most production and proprietary projects.

Quickstart:

pip install daytona

from daytona import Daytona, DaytonaConfig

config = DaytonaConfig(api_key="YOUR_API_KEY")
daytona = Daytona(config)
sandbox = daytona.create()
response = sandbox.process.code_run('print("Hello World")')

Requires a valid Daytona API key from the Daytona Dashboard and network access to the Daytona API endpoint.

Verify before relying:

- Performance characteristics and latency of sandbox creation and code execution.
- Scalability limits (concurrent sandboxes, resource quotas per account).
- Pricing model and cost implications of sandbox usage.
- Data residency and compliance certifications for the hosted infrastructure.

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 8.9M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

sandbox management sdk, secure code execution environment, isolated compute infrastructure, ai code execution platform, remote sandbox provisioning, sandbox-execution, ai-code-safety, infrastructure-as-code

[View on SkillFed](https://skillfed.io/packages/daytona) · [View on PyPI](https://pypi.org/project/daytona/)