趋势榜

YC (S26) | Open Computer History | 在本地持续录制屏幕，并向你的 Agent（Claude、Codex、Openclaw、Hermes、Runner 等）提供上下文。YC (S26) | Open Computer History | Record your screen continuously locally and provide context to your agents (Claude, Codex, Openclaw, Hermes, Runner...)

AI Agent 的通用记忆层。提供可扩展、可互操作的记忆存储与检索，为下一代自主系统简化 AI agent 状态管理。Universal memory layer for AI Agents. It provides scalable, extensible, and interoperable memory storage and retrieval to streamline AI agent state management for next-generation autonomous systems.

面向终端的开源编码 Agent，由社区集体而非公司构建。自带模型，代码保留在本地，不欠任何人。An open coding agent for your terminal, built by a community collective rather than a company. Bring your own model, keep your code on your machine, and owe nothing to anyone.

🎨 NeMo Data Designer：从零或从种子数据生成高质量合成数据。🎨 NeMo Data Designer: Generate high-quality synthetic data from scratch or from seed data.

为你的 AI agent 装上看 PDF 的眼睛——通过 MCP 提供结构化文本、表格、OCR、视觉证据和页级引用。原生 Rust，本地优先。Give your AI agent eyes for PDFs — structured text, tables, OCR, visual evidence, and page-level citations via MCP. Native Rust, local-first.

一个经过验证、社区维护的目录，涵盖 896 个使用 Jev 构建的开源项目。A verified, community-maintained catalog of 896 open-source projects built with Jev.

MIT 许可、无 COM 依赖的 .NET 库，用于创建、读取、编辑、转换、渲染和提取 Office、PDF、邮件、OneNote 及文本格式。MIT-licensed, COM-free .NET libraries for creating, reading, editing, converting, rendering, and extracting Office, PDF, email, OneNote, and text formats.

Awesome Jev：一份有来源支持的 TypeSafe System One 模型实地指南，包含 SDK、在线 demo、Agent 工具和独立评测。Awesome Jev: a source-backed field guide to TypeSafe's System One model, with SDKs, live demos, agent tools, and independent evaluations.

端到端人文学科写作助手——一个 Agent Skill（开放的 SKILL.md 格式）。提供 11 种模式，从苏格拉底式研究问题打磨到 AI 使用披露。中英双语（EN/中文），具备学科感知能力（文学/历史/哲学/艺术/宗教/语言学+跨学科）。四层批评、可校准的魔鬼辩护人、声音保留。End-to-end humanities writing assistant — an Agent Skill (open SKILL.md format). 11 modes from Socratic research-question sharpening through AI-use disclosure. Bilingual (EN/中文), discipline-aware (literature/history/philosophy/art/religion/linguistics + cross-disciplinary). Four-layer critique, calibratable devils advocate, voice preservation.

为 Hermes agents（同时支持 Claude Code 和 Codex）提供由 Jev 驱动的模型路由、记忆、压缩、Skill 选择、计算机与浏览器使用。Jev-powered model routing, memory, compaction, skill selection, computer and browser use for Hermes agents (also Claude Code and Codex)

A self-evolved, sovereign desktop AI agent built for real-world action — beyond model wrappers and coding copilots. / 真实实践中自我演化的独立桌面 AI Agent。

MCP 连接器，让你的 AI Agent 直接访问 typesafe ai 的 jev 模型mcp connector to give your AI agent direct access to typesafe ai's jev model

可复用 prompt 模板、多步工作流链路和质量门的 MCP 服务器。使用 operator 语法组合 Agentic 工作流；导出为原生 skill 到 Claude Code、Cursor、OpenCode 和 Gemini CLI。MCP server for reusable prompt templates, multi-step workflow chains, and quality gates. Compose agentic workflows with an operator syntax; export as native skills to Claude Code, Cursor, OpenCode, and Gemini CLI.

Codex 的每轮模型与推理路由，由 Jev（TypeSafe System One）驱动：为每一轮选择模型、思考深度和速度模式。Per-turn model & reasoning routing for Codex, driven by Jev (TypeSafe System One): picks the model, thinking depth and speed mode for every turn.

ChatCLI 是一个命令行应用，利用 LLM（如来自 Anthropic、OpenAI 等）实现终端内的交互式对话。其核心包含一个专用 Agent 和提供独特能力的工具。ChatCLI is a command-line application that utilizes LLMs—such as those from Anthropic, OpenAI, and others—to facilitate interactive conversations within the terminal. At its core, it features a specialized agent and tools that provide unique expertise.

本地 Codex 反向代理：轮换代理节点池、惰性健康故障转移、每模型 292 turn-state 采集/注入。支持 macOS + Windows。Local Codex reverse proxy: rotating proxy-node pool, lazy health failover, and per-model 292 turn-state collection/injection. macOS + Windows.

Android 上的仅接收 Wi-Fi 和 BLE 观测器。MIT 许可。Receive-only Wi-Fi and Bluetooth LE observer for Android. MIT.

F# 开发的 Sage Mode —— 支持解决方案或项目加载的 REPL、免费的 Live Testing、Hot Reload 以及会话管理。Sage Mode for F# development — REPL with solution or project loading, Live Testing for FREE, Hot Reload, and session management.

在 220k GB200 上训练前沿模型的完整脚本。"普罗米修斯从众神那里盗取火种，将其赐予人类。"Full script for training a frontier model on 220k GB200s. "Prometheus stole fire from the gods and gave it to man."

一切写作的单一应用the one app for everything you write

在旧款 Pixel 设备上解锁更新的 Google Pixel 相机功能。本项目将 Pixel 11 的新一代 Camera Looks 下放到 Pixel 6–10，并为 Pixel Camera 添加便捷的 10x 变焦开关。Unlock newer Google Pixel camera features on older Pixel devices. This project brings Pixel 11's next-generation Camera Looks to Pixel 6–10 and adds a convenient 10x zoom toggle to the Pixel Camera.

Apple Silicon 上的本地计算机使用：通过 macOS Accessibility API 驱动，由 laya 实现设备端类型化决策，由 CUA-S1-FORMS 实现表单填写。Local computer use on Apple silicon: on-device typed decisions with laya and form filling with CUA-S1-FORMS, driven through the macOS Accessibility API

Apple Silicon 上为任意 MLX 模型提供 Jev 风格的并行约束决策。单次前向即可输出符合 schema 的类型化 JSON。Jev-style parallel constrained decisions for any MLX model on Apple Silicon. Typed, schema-valid JSON in one forward pass.

AI 原生 OS。模型运行在自有硬件上，节点点对点联邦无需代理，API 兼容 OpenAI。可启动，自带 Wayland 合成器，8GB 即可运行。Apache 2.0。An AI-native OS. Models run on your own hardware, nodes federate peer-to-peer with no broker, and the API is OpenAI-compatible. Boots, has its own Wayland compositor, and runs on 8GB. Apache 2.0.

自托管、非托管的 Monero (XMR) 网页钱包：在官方 monero-wallet-rpc 之上的暗色响应式 UI。密钥保留在本地，发送采用两阶段并需显式确认手续费，提供 Windows 快速上手。Self-hosted, non-custodial Monero (XMR) web wallet: dark responsive UI on top of the official monero-wallet-rpc. Keys stay on your machine, two-phase sending with explicit fee review, Windows quick start.

一款能停下来向你询问的浏览器工具。无论是 2FA 验证码、CAPTCHA，还是只有你才能做出的选择：它会在屏幕上向你询问，然后在你已登录的标签页中继续执行。40 个工具，支持任意 MCP client，MIT 许可。The browser tool that can stop and ask you. A 2FA code, a CAPTCHA, a choice only you can make: it asks on your screen, then carries on in the tab you were already signed into. 40 tools, any MCP client, MIT.

让 Claude Code 和 Cursor 驱动 Chrome。浏览你的真实登录态：token -30%，成本 -25%，工具调用 -41%，工具定义 -34%，速度 +40%。直接 CDP，a11y-tree 引用，服务端计划执行器。MIT 许可，无付费档。Lets Claude Code and Cursor drive Chrome. Browse your real profile: -30% tokens, -25% cost, -41% tool calls, -34% tool defs, +40% faster. Direct CDP, a11y-tree refs, server-side plan executor. MIT, no paid tier.

混合神经符号 AI：Llama 3.2 1B + 精确数学 + 经验证事实，807 MB 即时 CPU 原生回答，无需 GPU。封装 .aef 分发。Hybrid neuro-symbolic AI: Llama 3.2 1B + exact math + verified facts — instant CPU-native answers in 807 MB, no GPU. Sealed .aef distribution