> ## Documentation Index
> Fetch the complete documentation index at: https://www.klavis.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Sandbox
> Scalable, isolated training and RL environments for real-world tool use
## The Problem
For LLM researchers, setting up LLM training or reinforcement learning environment for real-world tool use is complex and painful:
* Managing different environment or test accounts
* Implementing MCP Servers and handling various authentication issues
* Initializing realistic data
* Resetting states between multiple runs
* Ensuring isolation across concurrent sessions
## The Solution
Klavis MCP Sandbox as a Service solves these challenges. In addition to letting your model interact with our comprehensive MCP server ecosystem, you can use our sandbox infrastructure to easily **verify and reset data** on any concurrent run.
Our sandbox infrastructure is **horizontally scalable**, so it can handle any number of concurrent sessions as you need.
## Lifecycle
Request a sandbox based on the external services you need (Snowflake, GitHub, Notion, Woocommerce, etc.) and get an MCP server URL for that isolated instance.
Load a deterministic "world state" in JSON format or via API. We handle everything—creating databases, setting up data, and more.
Let your LLM / AI agent use MCP tools against the sandbox as if it were the real app. You can use multiple MCP servers with many tools simultaneously.
Access the full sandbox state to programmatically compare against your ground truth—whether your LLM completed the task correctly or not.
Wipe the sandbox back to a clean state and kick off the next run.
## Video
## Resources
Create sandboxes, seed data, run an agent, then verify and clean up.
Manage isolated sandbox environments for training/eval: pooling, init, export, teardown.
Use Klavis MCP Sandbox with Eval Protocol for model training and RL at scale.