Developer AlexWortega has introduced junoclaude—a specialized plugin for Claude Code that allows artificial intelligence to interact directly with the space simulator Juno: New Origins (formerly SimpleRockets 2) via MCP and HTTP protocols.
What Happened
The junoclaude plugin uses an MCP server and an HTTP bridge to transmit telemetry and flight controls in real time. This enables Claude to design rockets using the craft_build tool, write control programs in the Vizzy language, and analyze game logs directly within the simulator environment.
Context
The integration is implemented through the Model Context Protocol (MCP), which allows LLMs to access specific tools and data in real time. Using closed software environments through such custom protocols transforms complex simulators into interactive testing grounds for autonomous agents.
Why It Matters for the Industry
This case demonstrates the potential to turn gaming and engineering simulators into safe sandboxes for testing Agentic Workflows and Reinforcement Learning (RL). The development of similar MCP servers could lead to the standardization of interaction interfaces between the "brain" (LLM) and the "body" (simulator or robot) in automated design.
Why It Matters for Users
Users can delegate full-scale space design and flight programming tasks to Claude, effectively creating an autonomous "research engineer" within Juno: New Origins.
What Is Not Yet Known / Limitations
There is a need to evaluate the feasibility of transferring the methods demonstrated in this game simulator to real-world industrial modeling.
Sources
Author
Look at AI, Editorial Staff