GiuliaAPIDocumentationTry the agent demo

Getting started

Build with GiuliaAPI.

Radio network simulation tools for applications and AI agents. Start with a working example, then explore the guides and reference.

Choose your starting point

Connect an agent

MCP guide
Expose simulation tools to your client.

Overview

GiuliaAPI is designed for AI agents that need to inspect, modify, run, and evaluate radio network simulations. It provides a human-readable REST interface, machine-readable tool schemas, and an MCP endpoint for clients that support tool calling through the Model Context Protocol.

Diagram showing that a GiuliaAPI workspace contains simulations
REST API MCP tools Workspace isolation Usage counters
Recommended public model: create a simulation first. GiuliaAPI creates or reuses a workspace automatically, and later calls can use the returned sim_id.
Public deployment: the hosted endpoint is intended for agent interaction and experimentation, not high-throughput batch jobs.

Based on OpenGiuliaSLS

GiuliaAPI is built on top of the open-source OpenGiuliaSLS project. OpenGiuliaSLS provides the underlying radio network simulation engine; GiuliaAPI adds an agent-friendly service layer around it, including HTTP endpoints, MCP tool schemas, workspace lifecycle management, quotas, and usage instrumentation.

This separation is intentional: OpenGiuliaSLS remains the simulation core, while GiuliaAPI focuses on making that core usable by external agents, applications, scripts, and notebooks.

Core Concepts

Workspace

An isolated container for one task, workflow, or analysis. A workspace owns simulations, counters, quotas, and transient state.

Simulation

A configured radio scenario. Simulations can be created, cloned, modified, executed, queried, and deleted inside a workspace.

Episode

One realization of the simulated environment, such as user placement or traffic state. Many KPIs aggregate over episodes.

Step

A time or execution unit inside a run. Quotas often use episodes multiplied by steps as the resource unit.

Parameters

Engineering or UE values that agents can inspect and modify before running a simulation.

Results

KPI outputs from completed simulations, available as raw or aggregated data depending on the endpoint.

Access

Authentication and workspace selection

Python quickstart

Complete executable example

Agent integration

Connect through MCP

Lifecycle and limits

Workspaces, retention, and usage

Reference

REST endpoints and machine-readable schemas