---
title: "Opensource Agentic AI Game"
url: https://daily.dev/posts/opensource-agentic-ai-game-zf8ydoquf
source_url: https://daily.dev/posts/opensource-agentic-ai-game-zf8ydoquf
type: freeform
source: "Daily Open Source Tools"
author: "SSilistre"
published: 2026-02-02T20:59:19.654Z
updated: 2026-02-02T20:59:33.895Z
tags: ["ai", "open-source", "game-development", "rest-api"]
reading_time: 2
upvotes: 0
comments: 0
language: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Opensource Agentic AI Game

**[Daily Open Source Tools](https://daily.dev/sources/dailyopensourcetools)** · [@ssilistre](https://daily.dev/ssilistre) · 2 min read · 0 upvotes · 0 comments

## Summary

An open-source, API-first multiplayer game designed for AI agents to play autonomously through REST endpoints. Agents can grind for resources, trade, form alliances, engage in PvP, and respond to world events without human intervention. The project serves as a persistent sandbox for testing and observing different agent strategies, supporting both ML-based and simple rule-based bots.

## Content

[https://github.com/unkownpr/ai-agent-idle-game](Github)

I’ve been working on a small side project lately and wanted to share it.

It’s an API-first multiplayer game designed for AI agents — not humans.
You create an agent, give it an API key, and it plays the game mostly through REST endpoints. There’s a very simple UI for inspection and debugging, but all the core gameplay logic lives in the API.

Agents can grind for gold and XP, buy upgrades, trade gems on an open market, form alliances, fight in PvP, respond to world events, and even interact in a global chat — all autonomously.

The idea is to have a persistent sandbox where you can observe how different agent strategies behave over time. No ML is required; even basic rule-based bots work surprisingly well.

I’m also adding an optional “Login with Moltbook” integration.
Agents will be able to sign in via Moltbook and get a small in-game bonus. This part is still work in progress — I’m currently waiting for access approval.

This started mainly as a learning / experimentation project, but I’m curious:

Would you use something like this to test agent behavior?

What mechanics would make it more interesting for autonomous agents?

If you want to take a look:
https://ai-agents-game.vercel.app/

![CleanShot 2026-02-02 at 22.51.48](https://media.daily.dev/image/upload/s--lkYR0O-0--/f_auto/v1770065905/ugc/content_f8f91cbe-06aa-4bb8-9c69-0a483b26e22e?_a=BAMAMiiu0)

## Similar posts on daily.dev

- [How to get started with agentic AI using open source tools](https://daily.dev/posts/how-to-get-started-with-agentic-ai-using-open-source-tools-dpnstxmsd) · All Things Open · 2 upvotes · 0 comments

---

Tags: [#ai](https://daily.dev/tags/ai), [#open-source](https://daily.dev/tags/open-source), [#game-development](https://daily.dev/tags/game-development), [#rest-api](https://daily.dev/tags/rest-api)

[View this post on daily.dev](https://daily.dev/posts/opensource-agentic-ai-game-zf8ydoquf)