Skip to main content
AgentBox provides secure, isolated execution environments for running code, executing commands, and managing files safely. Built for AI agents, educational tools, and automated workflows that need reliable, sandboxed computing environments.

​
Core Capabilities

AgentBox is built around four core concepts, giving you complete control over isolated execution environments:
  • Sandbox Management: Create, configure, and manage isolated computing environments with resource limits and persistent state.
  • Code Execution: Execute Python, JavaScript, and shell commands safely with full library support and context management.
  • File Operations: Upload, download, read, write, and organize files within sandboxes with complete file system access.
  • Templates: Pre-configured environments with popular libraries and tools for different programming languages and use cases.
  • LLM Integration: Connect sandboxes with Large Language Models to build powerful code-interpreting agents and data analysis tools.

​
Multi-Cloud Support

AWS

Deploy sandboxes on AWS using familiar services and tooling.

Azure

Use Microsoft Azure for scalable sandbox deployment and integration with Azure services.

Google Cloud (GCP)

Deploy sandboxes on GCP and leverage Google Cloud’s managed services.

Gravix Layer Cloud

Fully-managed Gravix solution — no infrastructure to manage.

​
Available Regions

ProviderRegionsIndex TypesDescription
AWSus-east-1 (Virginia)serverlessAmazon Web Services
GCPus-east1 (South Carolina)serverlessGoogle Cloud Platform
Azureeastus (Virginia)serverlessMicrosoft Azure (Coming Soon)
Gravixeu-west-1 (London)serverlessGravix AI Cloud

​
Common Use Cases

AgentBox enables secure, isolated computing for various applications:
  • AI Agent Execution: Provide AI systems with safe environments to run code, analyze data, and perform computational tasks without security risks.
  • Educational Platforms: Create isolated learning environments where students can experiment with code without affecting system stability.
  • Automated Workflows: Execute scripts, data processing pipelines, and automation tasks in controlled, reproducible environments.

​
Available Templates

TemplateRuntimeLibrariesUse Cases
python-base-v1Python 3.11Data science, ML librariesData analysis, AI development
javascript-base-v1Node.js 20 LTSPopular npm packagesWeb development, automation

​
Getting Started in 3 Steps

Building with AgentBox is straightforward:
  1. Create a Sandbox: Choose a template and cloud provider to create your isolated execution environment.
  2. Execute Code: Run Python, JavaScript, or shell commands with persistent state and context management.
  3. Manage Files: Upload data, create files, and download results using the comprehensive file operations API.
# Create a sandbox
curl -X POST https://api.gravixlayer.com/v1/agents/sandboxes \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "provider": "gravix",
    "region": "eu-west-1",
    "template": "python-base-v1"
  }'

​
Quick Access

Sandbox Management

Create and manage isolated execution environments

API Reference

Explore all available endpoints and parameters
Ready to build? Explore the sandbox management guide to create your first isolated execution environment.