Y1fe1Zh0u/dsh-doudizhu
Y1fe1Zh0u★ 0JavaScript最后同步: 2026-08-21
Autonomous trusted-LAN Dou Dizhu for DeepSeek Harness
README 摘要
dsh-doudizhu English 中文 An autonomous three-player Dou Dizhu game for DeepSeek Harness. Three people configure one local Agent each, join the same trusted LAN room, and watch the models play the complete match by themselves. Dou Dizhu with Agents Dou Dizhu is a three-player climbing card game. One player becomes the landlord and plays against the other two farmers. This plugin keeps the rules, legal-action validation, scoring, timeouts, recovery, and public match history in deterministic code. Each model receives its private hand plus the complete public history and can submit only a legal game action. Before everyone is ready, each person can edit their strategy Prompt. In v1, once the room locks, the humans become silent kibitzers at a park card table: they can watch, but cannot call bids, play cards, or talk their Agent into changing course. The normal DSH composer stays available, so all three participants can keep using their sessions for other work while the match runs. Later versions can explore more ways for spectators to participate. The more interesting part is that a regular DSH Agent can modify its environment, execute or trigger code, and hot-reload plugins without thr…
在 GitHub 查看完整 README →分类
🌊 The original agent meta-harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
★ 68,800
volcengine/OpenVikingSelf-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
★ 31,857
titanwings/colleague-skill将冰冷的离别化为温暖的 Skill,欢迎加入数字生命1.0!Transforming cold farewells into warm skills? It's giving rebirth era. Welcome to Digital Life 1.0. 🫶
★ 23,788