---
id: langchain-daytona
version: "0.0.8"
license: MIT
license_treatment: permissive
maintenance: active
---

# langchain-daytona — Daytona sandbox integration for Deep Agents

License: permissive · Maintenance: active · Downloads: 260.2K/mo

## What it is and what it does

langchain-daytona provides a backend adapter that connects LangChain's Deep Agents framework to Daytona sandboxes, enabling agents to execute arbitrary code in isolated environments. The package wraps Daytona's sandbox API into a DaytonaSandbox class that accepts agent execution requests, manages timeouts, and polls for results.

The integration is lightweight—it depends only on daytona and deepagents—and targets modern Python versions (3.11+). It is intended for scenarios where an agent needs to run code safely without access to the host system, such as executing user-provided scripts or untrusted logic as part of an agentic workflow. The package is actively maintained and carries no known security vulnerabilities.

Use it for:

- Run user-submitted code within a Deep Agent workflow without exposing the host system to arbitrary execution.
- Execute generated or dynamically constructed code in an isolated Daytona sandbox as part of an agent's reasoning loop.
- Build multi-step agent pipelines where intermediate code generation and execution require sandboxed evaluation.
- Prototype or test agent behaviors that involve code execution in a controlled, time-limited environment.

## Worth the install?

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

Integrates Daytona sandboxes with LangChain's Deep Agents framework, allowing agents to execute code in isolated sandbox environments.

Yes, if you are building Deep Agents workflows that require code execution and want a straightforward Daytona integration. The package is actively maintained, has no known vulnerabilities, and carries a permissive MIT license. Install friction is low. However, verify that daytona and deepagents meet your infrastructure requirements before committing, as this package is young (first release 2026-02-05) and relatively niche.

## Install

pip install langchain-daytona
uv add langchain-daytona
poetry add langchain-daytona

## Installing langchain-daytona

Before you install: Low install friction with a pure-Python wheel distribution. Actively maintained as of 2026-08-14 with recent releases, though the package is young (first release 2026-02-05) and depends on daytona and deepagents.

License in practice: MIT license permits commercial and private use with minimal restrictions; you may use, modify, and distribute the package freely provided you retain the license notice.

Quickstart:

pip install langchain-daytona

from daytona import Daytona
from langchain_daytona import DaytonaSandbox

sandbox = Daytona().create()
backend = DaytonaSandbox(sandbox=sandbox, timeout=300, sync_polling_interval=0.25)
result = backend.execute("echo hello")
print(result.output)

Requires Python 3.11 or later; daytona and deepagents must be installed and accessible.

Verify before relying:

- How daytona and deepagents are installed and configured (whether they have their own system dependencies or setup requirements).
- Whether timeout and sync_polling_interval parameters are the primary configuration surface or if deeper customization is available.
- Performance characteristics and resource overhead of sandbox creation and code execution.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 260.2K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

daytona sandbox integration, langchain agent sandbox execution, isolated code execution environment, deep agents sandbox backend, langchain daytona integration, agent code sandbox, safe code execution langchain, agent-framework, sandbox-execution, code-isolation

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