Technical notes for E 2 B
Create, manage, and terminate secure cloud sandbox environments (lightweight Linux VMs) for executing AI-generated code. Run Python, JavaScript, and other code in isolated Jupyter-based interpreters with stateful sessions. Execute shell commands, install packages, and manage processes. Perform filesystem operations including creating, reading, writing, deleting, uploading, and downloading files. Pause and resume sandboxes to preserve full state including memory and running processes. Create snapshots of running sandboxes to rapidly spin up new instances from a known state. Build custom sandbox templates from Dockerfiles with configurable CPU, memory, and dependencies. Control desktop GUI environments with programmatic mouse, keyboard, screenshot, and streaming capabilities. Configure port forwarding for network access, connect external storage buckets, and manage MCP server integrations. Register webhooks for sandbox lifecycle events (created, updated, killed).
Frequently asked questions
Common questions about connecting E 2 B to AI agents with Metorial.
Can Metorial connect E 2 B to AI agents?
Yes. Metorial connects AI agents to E 2 B through a governed integration layer, so teams can use the provider while keeping access controlled and observable.Does the E 2 B integration work with MCP?
Metorial is MCP compatible and lets teams expose approved provider tools to MCP-capable agents and clients through a controlled access layer.How does Metorial control access to E 2 B?
Metorial applies policies across users, groups, providers, agents, and individual tools, then records the context around every agent interaction.Can teams trace E 2 B activity from agents?
Yes. Metorial records provider activity so teams can inspect tool calls, troubleshoot integrations, and give security teams the visibility they need.
