A local-first desktop workspace for persistent AI teammates. Run Codex, Claude, and Grok with dedicated workspaces, task queues, file sharing, browser control, and agent-to-agent collaboration.
SOMA 是一款开源桌面 CAD 应用：你用自然语言描述需要的零件，Claude Code、Codex CLI、OpenCode 等编程智能体用 Python（CadQuery）写出参数化模型，SOMA 实时构建并显示 3D 结果。适合需要外壳、支架、安装座的创客和 3D 打印爱好者，不会 CAD 也能上手，导出 STEP、STL、3MF 即可打印或加工。 亮点 不绑定任何 AI：使用你自己已有的 CLI 智能体和订阅，模型选择和权限都由你掌控 模型即代码：每个模型都是可读的 CadQuery 脚本，每一步都有 Git 历史，可预览和恢复 和智能体“指着说”：在 3D 视图中点选零件、精确测量、直接在画面上标注，再发送给智能体修改 不只是建模：剖切、爆炸视图、参数面板、标准件（螺钉、螺母、热熔螺母）、FEM 强度分析（CalculiX）、动画渲染和带尺寸的工程图（PDF、DXF） 开箱即用：Windows、macOS、Linux 安装包已内置 Python、CadQuery、Git 和 CalculiX，GPL-3.0 开源 （Note from the maintainer: this text was written with the help of an AI translation, as I don't read Chinese. Replies in English are easiest for me, Chinese is fine too.）
Hey HN, I'm Christian and this is a new OSS project I've been noodling on for the last couple of weeks. Hob is a free, MIT-licensed desktop app for Mac, Windows and Linux that works on your files the way the Claude and ChatGPT desktop apps do, except the model runs on your own computer. It uses open-weight models (currently Gemma 4 and gpt-oss). All your chats and files never leave your system. The only things that go online are web search (can be turned off in settings) and any MCP connectors you add yourself. Main features: - Automatically finds the best open-weight model for your hardware on first install (Mac/Linux/Windows) - Writes sandboxed Python code for ad hoc tools (data processing, document conversion, graphs etc.) - Built-in web search (powered by SearXNG) - MCP support - Image recognition - CRUD to your local filesystem - Works fully offline with no internet Before building this I tried every local AI app I could find, but it seemed each one had different problems. The main thing I noticed is that none of them could do what the ChatGPT and Claude apps do with your files, e.g. read and change files in a folder, look at images, run code in a sandbox, draw a chart. The most advanced thing any of them did was paste a plain text file into a small context window, which basically failed as soon as you had a file larger than a few KB. I couldn't find anything that could easily parse images and bigger PDFs. The OSS solutions that were more advanced all seemed to default to hosted models (either OSS or closed), which defeats the entire point of the privacy that on-device OSS offers. They also seemed to push towards an account/open-core style business. I build it using Electron bridged with llama.cpp in its own process. On initial load, the system will try and figure out if you have a compatible graphics card and if not default to CPU-only (I ran into a lot of issues of graphics card beeing seemingly compatible but then crashing on first query, there's a whole saga to this lol). Code execution happens in a sandbox that can only write to the chat's folder and has no network unless asked (sandbox-exec on macOS, Landlock on Linux and a low-integrity process on Windows). Images go to a small Qwen3-VL model. Dictation is a local Whisper model. The hardest part was definitely to get the smaller models to use tools reliably. Ran into a ton of issues of 4B models happily saying "here's your chart" without ever calling the chart tool or get stuck in an infinite call loop. I spent a fair amount of the work building guards vs tweaking prompts (that didn't really work for small models). I understand the pushback that local models are weaker than frontier cloud models. This is true, but for most tasks for most normal people, I'd argue you rarely need the newest frontier model. From my own evals, systems with 16+ GB of RAM do really well, and if you have a GPU or more RAM, you get very close to what you'll get out of the box with a frontier hosted model. Excited to hear what you think and see how it runs on other systems!
Know the risk of merging a pull request, and where it lives. Free, open-source CLI and GitHub Action: after your AI review, farol shows the riskiest files, what they reach, diff coverage and contract breaks. TypeScript/React, Go, Python.
Weblink 是我开发的开源 Windows 远程桌面工具，重点是低延迟交互和一对多桌面共享。受控电脑运行 Windows 客户端，其他设备通过浏览器加入同一房间即可观看桌面，获得授权后还能操作鼠标和键盘。 主要特点： 浏览器即用：手机或电脑上的控制端无需注册账号，也无需安装客户端。 一对多共享：多个设备可以同时观看同一个 Windows 桌面，适合远程演示和协作。 低延迟设计：采用 Windows 原生屏幕采集、受支持的硬件编码和 WebRTC；兼容的观看端可复用编码器。实际延迟取决于硬件与网络条件。 配套功能：支持文件传输、剪贴板共享、聊天及语音视频通话。 项目采用 MIT 协议，支持自行部署。连接需要信令服务，无法直连时可通过 TURN 中继。
MCP server that gives AI agents safe access to WhatsApp: read, receive, reply, media and reactions, with encrypted storage and phone numbers that never leave the server.
A Claude Code mod that holds subagent fan-out, heavy-context prompts and retry loops before they spend your usage window. It reads your plan's usage, counts agents across every session, and asks in Claude Code's own dialog.
文档： 开源 MCP 服务： 这是一次来自产品方的工具自荐。Sorsa API 提供公开 X（Twitter）数据的只读 REST 接口，包括用户资料、帖子、搜索、关注者与关注列表、列表、社区及互动验证，可用于数据分析、研究和 AI 应用。 使用一个 Sorsa API Key 即可接入，无需提供 X 账号密码或 Cookie，也不需要 X 开发者账号。提供在线 API Playground、OpenAPI 描述文件，以及可在 Cursor、Claude Desktop 等客户端使用的 MCP 服务。 新用户可试用 100 次请求，无需信用卡；更多用量需付费。服务不支持发帖、私信或读取私密账号内容。Sorsa API 是独立服务，与 X Corp. 无隶属关系。
Hi HN! We're releasing an open-source CLI for hillclimbing Agent Experience. That means optimizing your agent-facing surfaces (Skills, CLIs, SDKs, and docs) so that your users' coding agents can easily use your product. Pointing your own agent at the is enough to get started. https://github.com/gauge-sh/gauge-cli We've found that power users generally want to hand a CLI/MCP to their agents and get started from the terminal. They rarely want to log in and use the web UI themselves. We think developer tools must be agent-friendly to remain relevant. Here are some common findings that Gauge identifies: - agents get confused with CLI/SDK behavior and start source diving - interactive commands block agents - agents use deprecated/old CLI commands - agents don't use Skills when you'd expect - agents confidently assert limitations that don't exist - inconsistent terminology confuses agents - agents can't figure out how commands fit together into a workflow Locally, the Gauge CLI uses Docker containers to run agent harnesses (you can bring your own subscriptions). It finds where agents get confused or stuck and tests changes using your pre-release artifacts (Skills/CLI etc). We normalize events from Claude Code, Codex, Pi and OpenCode into standardized transcripts, and store them locally, alongside structured information in SQLite so that your agent can easily analyze the data. This enables automatic hillclimbing toward a great Agent Experience. We also offer hosted compute that enables more features, built on our custom Gateway (for credential brokering and transparent, per-session response rewriting). Hosted compute enables many more concurrent sessions, easier collaboration with teammates, a GitHub App for PR reviews, sessions with logged-in MCPs/APIs, and most importantly it lets you close your laptop while the sessions actually run. You don't need a Gauge account to run the full loop end-to-end though. We expect that local execution with the CLI is enough to find and fix several agent experience issues in a typical project. You'll be able to validate fixes for these issues locally, before publishing new versions (we inject pre-release Skills and code artifacts into the sessions). We would love your feedback!
