{"data":{"slug":"fuxiailab-civagent","name":"CivAgent","tagline":"LLM-based Human-like Agent acting as a Digital Player within Unciv","github_url":"https://github.com/fuxiAIlab/CivAgent","owner":"fuxiAIlab","repo":"CivAgent","owner_avatar_url":"https://avatars.githubusercontent.com/u/47130993?v=4","primary_language":"Python","stars":163,"forks":14,"topics":["aiagent","game","llm-agent","llm-evaluation"],"archived":false,"github_pushed_at":"2025-03-17T10:03:25+00:00","maintenance_label":"Dormant","url":"https://www.graphcanon.com/tools/fuxiailab-civagent","markdown_url":"https://www.graphcanon.com/tools/fuxiailab-civagent.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/fuxiailab-civagent","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=fuxiailab-civagent","description":"CivAgent is an LLM-based Human-like Agent acting as a Digital Player within the Strategy Game Unciv.","homepage_url":"https://github.com/yairm210/Unciv","license":"MPL-2.0","open_issues":4,"watchers":13,"ai_summary":"CivAgent is an AI agent that uses LLMs to act as a digital player in the strategy game Unciv, aimed at providing players with human-like opponents and facilitating research on AI agents.","readme_excerpt":"# CivAgent: Large Language Models based Human-like Agent\n\nEnglish | [中文](docs/README_chinese.md) |\n<p align=\"center\" width=\"100%\">\n\n\n\n**Moved the project repository to a new location: https://github.com/asdqsczser/CivAgent.**\n\nCivAgent is an LLM-based Human-like Agent acting as a Digital Player within the strategy game [Unciv](https://github.com/yairm210/Unciv).\nThe project aims to address the difficulty of players not being able to find human playmates,\nand seeks to construct a low-cost data flywheel to aid in the research of LLM-based Agents.\nBy deeply integrating Large Language Models (LLMs) with core gameplay,\nwe believe this is exactly an AI Native Game. All glory are attributed to Unciv.\n\n\n\nNative Unciv: https://github.com/yairm210/Unciv\n\nUnciv Wiki: https://civilization.fandom.com/wiki/Unciv\n\nGame Client Download (developer version, no Discord chat function): [Mac](https://drive.google.com/file/d/1Ohx6pvcdZbVzte0cAaXPHEAIRpJ2obqK/view?usp=sharing)   [Windows](https://drive.google.com/file/d/1ap99uZnhcpgIkDgJKAaPk277bTBi34Ag/view?usp=sharing)\n\nGame Client Download (player version, with Discord chat function, to be released in early August)\n\nUnciv We Used (source code): https://github.com/asdqsczser/Unciv/tree/release/fuxi_ver\n\nPaper: https://arxiv.org/pdf/2502.20807\n\nAppendix: https://github.com/fuxiAIlab/CivAgent/blob/main/paper_appendix.pdf\n\nYouTube: https://www.youtube.com/playlist?list=PL9G00-od8ezYNblsqMca7urrLoHSZ7Vs3\n\n## News\n **03/17/2025**: Moved the project repository to a new location: https://github.com/asdqsczser/CivAgent.\n\n**06/20/2024**: Code for benchmark reproduction (developer version) of research paper is open-sourced. The version for players is expected to be released in early July.\n\n**06/09/2024**: The paper is submitted to NeurIPS 2024 Track Datasets and Benchmarks, [Under Review](https://openreview.net/forum?id=AuT65qKLrr).\n\n\n## 👨‍💻 Developers\nPlease refer to [Documentation](/docs/README_for_developer_chinese.md).\n\n## 🙋 Contribution\nPlease refer to [Documentation](/docs/README_for_contribution.md).\n\n## ⚡ How to play\nThe AI service can only be used in the multiplayer game mode and is only supported on the Windows and Mac platform.\n\n\n**Note that your save files and chat data will be collected by the server (no personally identifiable data) and used for non-commercial purposes (AI effect improvement for this project).**\n### Step 1. Download the game client\nReleased in early July.\n\n\n### Step 2. Set the server address\nOpen the game, navigate to the 'Options' and 'Multiplayer' interface, adjust the synchronization frequency to 3 seconds, modify the server address to http://sl.office.fuxi.netease.com:44952 and test the link.\n\n### Step 3. Start a new game\nOpen the 'Start new game' interface, check the 'Online Multiplayer' option in the bottom left corner. It is recommended to begin with the 'Medieval era'. On the right, please click on 'Set current user', and select a civilization from the following civilizations: China, Mongolia, Egypt, Aztecs, Rome, or Greece.\n\n### Step 4. Copy the GameId\nAfter creating the game, the GameId will be automatically copied to your clipboard, or you can find the GameId of the game you created in the multiplayer game interface.\n\n### Step 5. Chat with civilizations on the Discord\n[CivAgent Discord](https://discord.com/channels/@me/1196286976639369297)\n\n### Step 6 (optional). Enhancing AI experience using LLMs such as GPT-4\nThe current default AI service utilizes the free large-scale model. For an enhanced gaming experience, you may consider subscribing to OPENAI's GPT-4 service.\n\n\n\n\n\n\n## 🙋 Motivation & Discussion\n### To Fans of Strategy Games\nIn strategy games, the experience of multiplayer gameplay is much better than that of computer AI, primarily due to the fact that human players make more rational strategic decisions, have a flexible diplomatic strategy, and can employ natural language for negotiation and deception.\nA typical example is when the emergence of a new conq","github_created_at":"2024-06-05T08:52:29+00:00","created_at":"2026-07-11T12:01:46.358906+00:00","updated_at":"2026-07-29T00:00:15.33752+00:00","categories":[{"slug":"ai-agents","name":"AI Agents","url":"https://www.graphcanon.com/categories/ai-agents","markdown_url":"https://www.graphcanon.com/categories/ai-agents.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/ai-agents"},{"slug":"evaluation-observability","name":"Evaluation & Observability","url":"https://www.graphcanon.com/categories/evaluation-observability","markdown_url":"https://www.graphcanon.com/categories/evaluation-observability.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/evaluation-observability"}],"tags":[{"slug":"aiagent","name":"aiagent"},{"slug":"game","name":"game"},{"slug":"llm-agent","name":"llm-agent"},{"slug":"llm-evaluation","name":"llm-evaluation"}],"trust":{"provenance":{"is_fork":false,"github_id":810730643,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-07-29T00:00:14.210Z","maintenance":{"label":"Dormant","score":18,"methodology":"github_public_v1","releases_90d":0,"days_since_push":498,"last_release_at":null},"security_summary":{"status":"findings","scanner":"osv@v1","low_count":43,"high_count":0,"last_scan_at":"2026-07-11T12:01:52.612Z","medium_count":0,"scan_profile":"deps","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-07-29T00:00:14.696Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-07-29T00:00:14.696Z"},"license_spdx":{"value":"MPL-2.0","source":"github.license","observed_at":"2026-07-29T00:00:14.696Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":{"notes":["The project relies heavily on the Unciv game engine for functionality.","A non-commercial data collection policy is in place to improve AI effects within the framework of this specific research."],"requires_docker":false},"constraints":{"requires_docker":false},"when_to_use":["CivAgent is ideal for those involved in research on AI agents, particularly focusing on integrating LLMs with gameplay, as it uses these models deeply embedded within Unciv's core mechanics.","It suits users who are seeking a challenging, human-like opponent while playing the strategy game Unciv and are interested in contributing to AI advancements through non-commercial data collection."],"when_not_to_use":["CivAgent is not suitable for players looking for immediate commercial-grade gaming experiences due to its specific focus on research and potentially lower accessibility.","It may not be ideal for those seeking direct interaction with state-of-the-art language models like GPT-4, as the current default relies on free large-scale models.","Users should avoid CivAgent if they require support beyond Windows or Mac platforms, as it is currently only supported on these systems."],"source":"enrich:decision_facts","observed_at":"2026-07-12T14:09:51.158Z"},"constraint_facets":{"requires_docker":false},"decision_summary":[{"label":"Requirements","value":"The project relies heavily on the Unciv game engine for functionality.; A non-commercial data collection policy is in place to improve AI effects within the framework of this specific research."},{"label":"Adopt for","value":"CivAgent is an AI agent that uses Large Language Models (LLMs) to act as a digital player in the strategy game Unciv, designed for research purposes and to provide human-like opponents."},{"label":"License detail","value":"CivAgent operates under the MPL-2.0 license, offering permissive terms for use in both open-source and proprietary applications."}]}}