📈 GH Trending Explorer 🎲 Surprise me
jokercaijojo / bianbian-maixiaoshu

百变麦小薯 —— 会变身的麦当劳优惠蹲点机器人 | 麦当劳程序员创意开发大赛参赛作品

▲ 65/7d ★ 82 Python 7d ago
mozihe / z-localization

2026年 东北大学T-DT实验室 RoboMaster超级对抗赛 哨兵定位开源

▲ 65/7d ★ 76 C++ 3d ago
AIDotNet / TKCode.Arc

Arc-style (uiarc.dev) component library for Avalonia, built on AtomUI

▲ 65/7d ★ 73 C# today
17-log / mc-keeps

麦味拾光:照片手帐、贴纸回顾与麦当劳 MCP 只读订单关联。日月年回看,数据留在本机。

▲ 65/7d ★ 72 HTML 3d ago
Metamorfolog / file_action

Click-free file conversion for Windows and Linux: drag a file, tap a key, drop it on a radial wheel. Runs locally.

▲ 65/7d ★ 65 C# 3d ago
eqyvlbekmlqq / tiktok-view-bot

TikTok viewer bot: multi-threaded view generator, proxy rotation, watch-time randomization, captcha bypass, live viewer count inflation

▲ 65/7d ★ 65 HTML today
GLMap / earstrip

Fast polygon triangulation into triangle strips: ear clipping with a constrained Delaunay fallback for broken rings. Header-only C++17 and WebAssembly, with an earcut-compatible JavaScript call.

▲ 65/7d ★ 65 C++ 1d ago
TriwindBryan / fulton

fulton - A launcher for decomp, recomp and remake PC ports of classic games, for Windows and Mac.

▲ 64/7d ★ 81 JavaScript 2d ago
ajogalekar / claude-codex-loop

Autonomous Claude Code execution with hourly Codex review, reconciliation, and bounded follow-up prompts.

▲ 64/7d ★ 66 Python 7d ago
urdtbittefnu / deadlock-external

Deadlock external cheat: ESP wallhack, aimbot, skeleton overlay, radar — educational reverse-engineering reference, injector included.

▲ 64/7d ★ 64 Python today
852wa / JIZURA

歌詞から文字PVを自動で組み立てるブラウザアプリ

▲ ~63/7d ★ 2.1k HTML 24d ago
Mak5er / AirCard-iOS

Apple Wallet card skins and lock screen passcode themes on iOS 27.

▲ ~63/7d ★ 1.9k Swift 19d ago
ruc-datalab / EvoOntology

EvoOntology: A Self-Evolving Ontology Layer for Data Agents ⚙️ EvoOntology插件为Claude Code/Codex 建立&进化本体层

▲ ~63/7d ★ 1.0k Python 15d ago
abhi245y / PS5LM-Plus

llama.cpp on a jailbroken PS5: run any GGUF model natively on the console, starting with Qwen 3.8

▲ ~63/7d ★ 459 C++ 8d ago
himovo / movo

Turn DeepSeek Harness into a self-hosted enterprise Agent platform with knowledge, deep research, content generation, vibe coding, browser automation, governance, and admin controls.

▲ ~63/7d ★ 357 Python today
AhernResearch / grava

GRAVA: A Unified Grounded Reasoning-to-Action Framework for Autonomous Driving. Project overview.

▲ ~63/7d ★ 355 21d ago
mikamikasuki / forest

FOREST is a research agent for long-horizon science, where evidence gets lost, errors compound, and goals drift. It explores ideas, reads literature, forms hypotheses, runs real experiments, analyzes results, revises paths, and builds a full paper—all in an editable graph you can inspect, branch, redirect, with claims and figures linked to evidence

▲ ~63/7d ★ 345 Python 1d ago
FishBottle7 / opencode2dsh

DSH plugin — free OpenCode Zen models for DeepSeek Harness (DSH). Free LLM API, no API key needed. 在 DSH 中使用 OpenCode Zen 免费模型,无需 API key

▲ ~63/7d ★ 342 TypeScript 4d ago
acryldev / acryl

ACRYL - Agent Context Relay Yielding Lifecycles. One persistent workspace, one canonical context, any coding agent.

▲ ~63/7d ★ 298 TypeScript today
evanbaker315 / AttentionOS

Star the repo to help support! The open source operating system for creating viral content on social media.

▲ ~63/7d ★ 153 TypeScript 3d ago
heyongjun666 / AICU-Graph

可扩展到各个领域的多 Agent 协作平台:自动分析任务,动态组建由不同角色组成的 AI 团队,协同完成复杂任务。

▲ ~63/7d ★ 146 Python 2d ago
jaikydota / BullFlow

推量牛 BullFlow — 开源企业级 RAG 解决方案:从企业知识库、营销内容自动化生产到全域内容分发 | Open-source enterprise RAG solution: knowledge base → automated marketing content → omnichannel publishing. 知识库全链路本地开源模型,无需付费模型;AI Agent 创作图文 / 图片 / 视频,一键发布抖音、小红书、公众号、LinkedIn、YouTube 等 17 个平台。Commercial-grade, multi-tenant, billing & admin built in. MIT

▲ ~63/7d ★ 129 TypeScript 11d ago
r2m-ai / load-bearing-cot

Perturb-and-continue causal tests of chain-of-thought: does the written reasoning actually constrain the answer? Code, data, and probes for "From Decorative to Load-Bearing: Task Difficulty Shapes the Causal Role of Chain-of-Thought" (TMLR 2026).

▲ ~63/7d ★ 121 Python 8d ago
drakolordx7 / KillzoneRecomp

Native PC port of Killzone (PS2) via static recompilation. Mouse and keyboard, 120+ fps, upscaling. Bring your own disc.

▲ ~63/7d ★ 119 C++ today
michaelshimeles / bible-rs

A very fast native Bible reader for the Mac, written in Rust: typeset Scripture, instant search, highlights, notes, bookmarks. Public-domain translations built in; bring your own licensed ones.

▲ ~63/7d ★ 107 Rust today
agentbaseline / agentbaseline

Six security outcomes an enterprise must achieve to run AI agents, and the controls that evidence them. A draft for public comment.

▲ ~63/7d ★ 106 Python 11d ago
HyxiaoGe / fusion

面向多模型与 Agent 工作流的 AI 聊天产品,采用 Next.js + FastAPI monorepo 架构。

▲ ~63/7d ★ 98 Python today
zhu-hailin / chaoxing-ai-assistant

超星学习通辅助工具。可防暂停视频、章节测验AI一键填入答案、全自动刷课、课程作业AI一键完成(可识别图片)。

▲ ~63/7d ★ 96 JavaScript 1d ago
akira-init1 / ChannelTerm

Shared terminal sessions for humans and AI agents, starting with serial and hardware debugging.

▲ ~63/7d ★ 83 Go today
Rocky-true / vmware-to-kubernetes

Kubernetes 自动化运维与迁移工具:阿里云 3 主 HA、Compose→K8s、Terraform、Kubekey、部署验证与故障排查。Kubernetes automation, migration and ops toolkit.

▲ 63/7d ★ 80 Python 3d ago
5hyuns / maya-workflow-kit-for-blender

Maya Workflow Kit: the whole Maya workflow in Blender, in one install. Hotbox, marking menus, Channel Box, Attribute Editor, menu sets, shelves and Maya concepts. Free, GPL Blender extension for Maya users.

▲ 63/7d ★ 77 Python today
JenniceXie / sota-options-rl

SOTA: strategy-level reinforcement learning for options portfolio management — code release for the paper

▲ ~63/7d ★ 75 Python 17d ago
xiaoshe1337-star / Xsdsh-purge

沿用了此项目:https://github.com/YuJunZhiXue/dsh-purge 专注于deepseek harness破甲|逆向|渗透| 适配最新版deepseek harness :deepseek v4.1 flash 最新版

▲ 63/7d ★ 71 JavaScript 6d ago
lululynne / maimai-xianpin

麦麦现拼:到店自取、同店陌生人现场拼一个大套餐(麦当劳 MCP 开发者大赛)

▲ 63/7d ★ 63 TypeScript 6d ago
meta-pytorch / dist_moe

Dist-MoE implements Speed-of-Light distributed Mixture-of-Experts forward and backward kernels. They are highly optimized and fused comm-compute kernels for training and inference on SM100+ GPUs

▲ ~63/7d ★ 62 Python today
ASEbrahim / MaliBridge

Play Steam and PC games on Android with a Mali GPU. Works well on the Samsung Galaxy Tab S10+ and Galaxy Tab S11 Ultra; other Mali phones and tablets (MediaTek Dimensity, Google Tensor, Xiaomi XRing) are being tested. A Vulkan bridge for DroidDeck.

▲ ~63/7d ★ 61 C 1d ago
LarryE135 / dsh-flash-presets

降低Deepseek v4.1flash的Token消耗/Raising Token efficiency with same performance(Works for v4.1flash)

▲ ~63/7d ★ 60 JavaScript 1d ago
beyondtahir / beyondartist

AI design skill for freelancers, content creators and digital marketers. Think like a designer. Execute like an artist.

▲ ~63/7d ★ 52 Python 10d ago

Load more →