A Survey on Large Language Model-Based Game Agents (ACM CSUR)
-
Updated
Jun 7, 2026
A Survey on Large Language Model-Based Game Agents (ACM CSUR)
A local PTCG practice and rules-simulation project built with Godot 4.6 and GDScript.
[ICML 2026] GameVerse: Can Vision-Language Models Learn from Video-based Reflection?
🕵 Code for our EMNLP 2025 Main paper: "FlashAdventure: A Benchmark for GUI Agents Solving Full Story Arcs in Diverse Adventure Games"
LLM-powered AI agent framework for 4X mobile strategy games (Rise of Kingdoms, Evony, Lords Mobile, etc.). 5-layer hybrid architecture: OCR → FSM → World Model → Workflows → LLM Vision. 99% cheaper than pure LLM approach.
XOShift AI agent built with Python using minimax, alpha-beta pruning, iterative deepening, and heuristic board evaluation.
Ollama Support for Neocortex Unity SDK
GAT — Godot Agent Team: a full-cycle (design→implementation) Godot game-development agent / 游戏制作智能体. Multi-agent, style-locked asset generation, self-evolving knowledge base; runs under both Claude Code and Codex.
Deterministic, resource-first Arena Hero agent with hierarchical threat response, safe unattended deployment, and optional offline AI review
AI agent that plays Fishing Frenzy autonomously via Claude Code. Install, play, compete.
InstaFlow is an AI-powered Instagram management platform that helps users automate their Instagram activities using an intelligent agent. The application enables scheduling posts, analyzing engagement metrics, and leveraging AI to make data-driven decisions for Instagram content strategy.
A hierarchical fog-of-war strategy agent for Lux AI Season 3
Kaggle Orbit Wars evaluation toolkit and competition retrospective with deterministic matches, replay analysis, and safe agent packaging.
Resumable, screen-only Codex computer-use arena for locally installed Steam games
Un set de unelte pentru organizarea de partide de Splendor între programe
AI agent that plays 2048 using expectimax search and heuristic evaluation
CityU - CS4386 - AI Game Programming
An LLM-based game agent plays a classic puzzle game.
OMNIDEX for Virtuals Protocol GAME agents — discover, rug-check, quote, and swap Solana tokens non-custodially. Wraps the Autonomous Intelligence agent rail.
To associate your repository with the game-agent topic, visit your repo's landing page and select "manage topics."