What problem does it solve? Executing untrusted or LLM-generated code safely requires isolated environments, and configuring Cloudflare's Sandbox SDK involves precise wrangler.jsonc settings, Durable Object bindings, and API choices that are easy to get wrong. ## Core Features & Use Cases - Sandbox Lifecycle Management: Create isolated containers with getSandbox(), control sleep behavior, and destroy instances to free resources. - Command and Code Execution: Run shell commands with exec() or execute LLM-generated code with runCode() supporting Python, JavaScript, and TypeScript with persistent contexts. - File and Port Operations: Read, write, and list files in the sandbox filesystem, and expose HTTP services via preview URLs. - Use Case: Build an AI code interpreter where an agent writes Python analysis code, executes it in an isolated sandbox with pandas, and returns rich outputs like charts and tables to the user. ## Quick Start Ask the agent to set up a Cloudflare Worker using the Sandbox SDK that runs Python code in an isolated sandbox and returns the execution results.