yxc20089yxc20089Communityยท1 Agent Skills Included

OpenRA-RL

Play Red Alert RTS with code-controlled game commands

Lets language models and scripted bots play Command & Conquer: Red Alert through 48 game-control commands. Handles base building, resource gathering, army training, scouting, and combat without manual gameplay. Runs the game server in Docker with replay recording, benchmarking, and leaderboard submission built in.
npx skills add yxc20089/OpenRA-RL --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install OpenRA-RL?โ–ผ

Run `npx skills add yxc20089/OpenRA-RL --all -g -y` in your terminal to install this skill globally. You also need Docker installed, since the game server runs in a container.

How to make an AI play Red Alert?โ–ผ

Start the game server with `openra-rl server start`, connect the MCP server to your client, then ask your agent to start a game. It uses 48 commands to build bases, train armies, and attack enemies.

Does OpenRA-RL work with Claude Desktop and OpenClaw?โ–ผ

Yes. Run `openra-rl mcp-server` and add it to your MCP client config to expose all game commands as standard MCP tools.

Can I use local models with OpenRA-RL?โ–ผ

Yes. It supports local model servers like Ollama and LM Studio, plus cloud providers like OpenRouter, through a simple YAML config or CLI flags.

How do I watch replays of my agent's games?โ–ผ

Replays are saved automatically after each game. Run `openra-rl replay watch` to view the latest match in your browser via the built-in viewer.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’