趋势榜

全球首个开源的企业级世界模型。World's first open-source enterprise world model.

用 Rust 编写的高性能 AI 编码 Agent CLI，零 unsafe 代码High-performance AI coding agent CLI written in Rust with zero unsafe code

开源、local-first 的 Granola/Otter 替代方案，可被 Claude Code、Codex、Cursor 及任何 MCP 客户端查询。会议、通话与语音备忘录在本地转写为你拥有的 Markdown。Open-source, local-first Granola/Otter alternative that Claude Code, Codex, Cursor, and any MCP client can query. Meetings, calls, and voice memos transcribed on-device into markdown you own.

统一的 TUI 和 CLI，用于跨 11+ 个提供商索引和搜索你的本地 coding Agent 会话历史（Codex、Claude、Gemini、Cursor、Aider 等）。Unified TUI and CLI to index and search your local coding agent session history across 11+ providers (Codex, Claude, Gemini, Cursor, Aider, etc.)

阻止你的 AI 胡编乱造——AI 提出方案，确定性工具做决策，每条断言都对照真实证据进行核验。可信事实与上下文在重置后仍能保留。逆向工程作为验证场。MCP server + CLI。Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is the proving ground. MCP server + CLI.

使用开源 LLM 摄取文件用于检索增强生成（RAG），无需第三方，数据不出你的网络。Ingest files for retrieval augmented generation (RAG) with open-source Large Language Models (LLMs), all without 3rd parties or sensitive data leaving your network.

逐步学习 LLM 推理工程——从 KV cache、PagedAttention 和连续批处理，到 vLLM、SGLang 和 GPU。Learn LLM Inference Engineering step by step - from KV cache, PagedAttention, and continuous batching to vLLM, SGLang, and GPUs.

DLSS 5 神经视频与图像增强器DLSS 5 Neural Video & Image Enhancer

在 DirectX 11 和 Vulkan 游戏中实现 DLSS 5 Neural Rendering，也可通过 NVIDIA Optical Flow 在不支持 DLSS 的游戏中使用（画质较低）。一个 ReShade 插件。DLSS 5 Neural Rendering in DirectX 11 and Vulkan games, and optionally in games with no DLSS via NVIDIA Optical Flow (lower quality). A ReShade add-on.

Pragmatic AI Labs MCP Agent Toolkit —— 一个 MCP Server，用于让 Agent 编程更加确定性Pragmatic AI Labs MCP Agent Toolkit - An MCP Server designed to make code with agents more deterministic

适用于所有 PC 游戏（DX11/DX12/Vulkan）及所有 NVIDIA GeForce RTX 20、30、40、50 系列的通用 Steam 风格 Game Center 与一键式 Neural DLSS 5 注入器。具备 100% 原生 DLAA Feeder 模式、OptiScaler Bridge 和 Direct Injection。Universal Steam-style Game Center & 1-Click Neural DLSS 5 Injector for ALL PC Games (DX11/DX12/Vulkan) & All NVIDIA GeForce RTX 20, 30, 40 & 50 Series. Featuring 100% Native DLAA Feeder Mode, OptiScaler Bridge & Direct Injection.

领取专属于你的 cool'runs-on.dev' 子域名，永久免费使用。Grab your own cool'runs-on.dev' subdomain to use for free forever.

Pictaria Frame 的自托管配套服务器，由你的 Immich 媒体库提供支持。Self-hosted companion server for Pictaria Frame, powered by your Immich library.

在 Windows 上使用 Omarchy Linux，无需任何麻烦。Use Omarchy Linux on Windows without any hassle.

用 Python 绕过 Cloudflare 验证（Just a moment / 5s），通过 Peak API 获取有效的 cf_clearance cookie，支持无头浏览器与代理。Bypass the Cloudflare challenge (Just a moment / 5s) in Python and get a valid cf_clearance cookie via the Peak API - headless-friendly, proxy-aware.

Intent as Code｜AI 的工作流语言。一个文件、4 个动词、一个 Rust 二进制程序。本地优先，支持任意模型，AGPL-3.0。🦋Intent as Code | the workflow language for AI. One file, 4 verbs, one Rust binary. Local-first, any model, AGPL-3.0. 🦋

面向挪威政府的 Copilot、Agent、MCP、Skill 及编码相关 AI 功能综述All things Copilot, Agents, MCPs, Skills, and coding related AI features for the Norwegian Government

JabRef 用户文档。User documentation of JabRef

面向 UFC（巴西联邦大学）学术工作的社区化、现代化 LaTeX 模板。Template LaTeX comunitário e modernizado para trabalhos acadêmicos da Universidade Federal do Ceará (UFC).

俄勒冈州立大学研究生课程 Scientific Writing 的教学材料。Teaching materials for Scientific Writing, a graduate course at Oregon State University

查找你的 AI coding Agent 泄露到磁盘的敏感信息。免费、MIT 许可、零依赖、零网络调用。在我们公开发布的基准测试中击败 TruffleHog 和 GitGuardian 的引擎。Find secrets your AI coding agent leaked to disk. Free, MIT, zero deps, zero network calls. Beat TruffleHog and GitGuardian's engine on our own published benchmark.

用于文献综述的本地优先 Chrome 扩展：保存文章、生成 AI 摘要、与 Gemini 进行有依据的对话、记录丰富笔记，并导出带样式的 Excel 文件。A local-first Chrome extension for literature reviews: save articles, AI summaries, grounded chat with Gemini, rich notes, and styled Excel exports.

该仓库包含复现一项生态、进化与环境科学领域荟萃分析 meta-research 研究所需的全部代码与数据This is a repository that contains all code and data needed to reproduce the results of a meta-research study on meta-analyses published in ecology, evolution and environmental sciences

Question Spec：研究问题的选择契约，配套 qspec 工具和 RealTimeX 插件。与 Paperforge 配对使用。Question Spec: a selection contract for research questions, with the qspec tool and RealTimeX plugin. Pairs with Paperforge.

研究问题、论文、研究发现和实验。Research questions, papers, findings, and experiments

Per：按 vault 隔离的本地优先引文管理器，灵感来自 Mendely 和 Zotero。Per vault/local-first citation manager inspired by Mendely and Zotero

用纯 Markdown 追踪研究问题、假设与结论，便于 Agent 与团队成员查看尝试过程与结果。Google Open Knowledge Format 的实践资料。Track research questions, hypotheses and verdicts in plain markdown, so agents and teammates can see what you tried and what happened. A profile of Google's Open Knowledge Format.

适用于大学作业的易用 LaTeX 模板，支持数学、代码、图片与 D2 图表Accessible LaTeX template for university homework with math, code, images, and D2 diagrams

使用客户调查数据进行跑步机购买者画像统计分析，采用假设检验（t 检验、ANOVA）。Statistical analysis of treadmill purchaser profiles using hypothesis testing (t-tests, ANOVA) on customer survey data

面向英语专业本科生的一站式学术研究与论文写作工作流：文献精读、Source Map、Research Gap、语言学 APA 与文学 MLA。An end-to-end research and academic writing workflow for English majors.

用于人工审阅 evidence-backed WAKG 文献抽取结果的临时公开预览。Temporary public preview for human review of evidence-backed WAKG literature extraction.

面向基于丙泊酚的 ICU 镇静的计算数字孪生：SQL 数据提取、特征工程、消融实验和论文图表。Computational digital twin for propofol-based ICU sedation: SQL extraction, feature engineering, ablation experiments, and publication figures.

用于自动化论文分析与文献综述的 Agentic AIAgentic AI for Automated Research Paper Analysis and Literature Review

面向护理和生物医学研究的 Claude Skills——APA-7 审核、PRISMA-2020 系统综述、混合方法设计、稿件同行评审、执照考试推理，以及无标记动作捕捉人体工学。Claude Skills for nursing and biomedical research — APA-7 auditing, PRISMA-2020 systematic reviews, mixed-methods design, manuscript peer review, licensing-exam reasoning, and markerless motion-capture ergonomics.

基于 LLM 的 AI 研究 Agent：检索 arXiv、生成假设，以及一个能够拒绝自身输出的 Critic Agent。支持通过 Ollama 在本地运行，或使用 Anthropic API。LLM-backed AI research agent: arXiv retrieval, hypothesis generation, and a Critic agent that can reject its own output. Runs locally with Ollama or via the Anthropic API.