Start with the job, then find an open-source answer

claude-code-tools brings practical command-line utilities and Claude Code plugins into one project for Claude Code, Codex CLI, and similar agents. Users can add session-management, terminal-automation, or document-workflow capabilities as needed; Google Docs and Sheets are optional extras, and the search component requires a separate binary.

AppPorts moves large applications out of `/Applications` onto an SSD, SD card, or network storage while leaving a small launcher stub in the original location. Apps can still appear and launch normally through Finder and Launchpad, and the utility selects different strategies for native apps, self-updating software, suites, and supported App Store apps. It can also flag broken links when a drive is unavailable. System apps and running apps are blocked from migration, and the current download is not signed with an Apple Developer ID, so macOS may require manual quarantine removal.

Colorpicker is an Electron desktop tool for macOS, Windows, and Linux that samples colors from anywhere on the screen and saves or categorizes reusable values in Colorsbook. Pinning, shade bars, focus mode, random colors, and preferences support day-to-day color work. Clipboard color detection is still under development and currently recognizes only the first hexadecimal or RGB value it finds.

Quasar Awesome is a curated directory for developers working with the Quasar framework. It groups official documentation and community links with tutorials, video courses, app extensions, third-party components, development tools, and projects built with Quasar. The list is useful for learning and package discovery, but many external entries are neither created nor maintained by the Quasar team, and their compatibility may target different Quasar versions.

Natively is a desktop meeting, lecture, and interview assistant that captures system and microphone audio, performs on-device speech-to-text, and uses either bring-your-own model keys or a local Ollama instance for notes, answers, and searchable memory. Its separate audio channels are designed to keep speaker and user input distinct, while transcripts, embeddings, and credentials remain on the local machine by default. The source is available under a personal-use license covering personal, educational, research, and non-commercial use; commercial use requires a separate written license.

Deep Code CLI is a terminal coding assistant optimized for deepseek-v4. Run deepcode in a project to use adjustable reasoning depth, conversation resume, code or chat undo, image pasting, Agent Skills, and MCP integrations; its settings can also be shared with the VS Code extension. Other OpenAI-compatible models are supported, but the harness, tool constraints, and context behavior are designed primarily around DeepSeek.

PyGPT brings cloud providers and local Ollama models into one desktop assistant for Windows, macOS, and Linux. It supports chat, document retrieval, voice interaction, and agent workflows through an extensible plugin system. Some plugins can read or write files and execute generated Python or system commands, so users should understand and tightly control tool permissions before enabling host execution.

Open Cowork is a graphical AI agent workspace for Windows and macOS that can use Claude, OpenAI-compatible APIs, and several other model providers. Within a selected workspace, an agent can manage files, create PPTX, DOCX, PDF, and XLSX outputs through skills, connect to services such as browsers or Notion through MCP, and receive remote instructions through Lark or Slack. Basic protection restricts file access by path; WSL2 on Windows or Lima on macOS can add VM isolation. Users should still review permissions carefully because authorized file changes or deletions carry real risk.

Claude Code History Viewer reads local conversation data from many coding assistants, including Claude Code, Codex, Cursor, Gemini CLI, and OpenCode, and presents it in one offline interface. You can search across projects, compare providers, and review usage analytics. Desktop builds cover macOS, Windows, and Linux, while headless mode can export individual sessions as HTML or JSON; a full backup action archives all Claude Code sessions before automatic cleanup removes them.

Agent Skill Creator converts prose, PDFs, links, scripts, or transcripts into a reusable skill with a SKILL.md, working code, an evaluation specification, and an installer. It validates and security-scans the generated package, then adapts it for Claude Code, Codex, Cursor, Gemini, and other supported tools. This is aimed at teams that want to package repeatable internal workflows without hand-writing each skill.

AIaW is a local-first AI chat client with multiple workspaces, support for several model providers, MCP connections, plugins, reusable prompts, and editable Artifacts. Conversation data is stored locally for offline access, with optional cloud synchronization after sign-in across Windows, Linux, macOS, Android, and the web PWA. The project is deprecated and has been replaced by the rewritten Nya AI client, so existing users are directed to migrate rather than adopt AIaW as a new long-term setup.

@polkadot/apps provides a visual portal for exploring and interacting with supported Polkadot and Substrate chains. Users can open the hosted web application or run it through IPFS, Docker, or a local node; the desktop edition offers the same general interface while storing encrypted accounts on the local filesystem by default. Chain developers can customize API types, settings, and branding through the repository's configuration package.

Use an endless SVG canvas as a simple desktop whiteboard for diagrams, visual notes, and free-form idea mapping. The app supports panning and zooming, arrows, rectangles, ellipses, selection, copy and paste, folders, local file persistence, basic touch input, and light or dark themes. Individual canvases can be exported as PNG, JPEG, or SVG, and prebuilt releases are available alongside an Arch Linux package. A text tool is still unfinished, so the current workflow is centered on shapes and drawing rather than typed annotations.

LibVLCSharp exposes VideoLAN's LibVLC engine to .NET and Mono applications across desktop, mobile, server, and Unity environments. Developers can render video, output audio, handle files and network streams, and use encoding or streaming capabilities through a common API, with integrations for frameworks such as WPF, WinForms, MAUI, Avalonia, and Xamarin. Each target also needs the platform-specific native LibVLC package, because the bindings do not include the media engine by themselves.

Cate is a desktop coding workspace that places editors, terminals, browser tabs, documents, Git tools, and AI-agent chats on a zoomable spatial canvas. Panels can dock, split, detach, and persist across projects, while the built-in agent can connect to several model providers. Prebuilt releases are available for macOS, Windows, and Linux.

Elixir Desktop lets Phoenix LiveView developers build native-like local applications for Windows, macOS, Linux, iOS, and Android. It routes window, webview, menu, and notification operations through platform backends, using wxWidgets on desktop, a JSON bridge on mobile, or a browser backend for headless environments. An example application and setup guides provide a starting point. Core desktop and mobile execution works, but complete desktop installer generation, signing, and one-command cross-platform packaging remain roadmap items.

poi is a desktop browser and toolkit for Kantai Collection that combines game access, caching, data synthesis and analysis, notifications, and a plugin system in one application. It supports HTTP and SOCKS5 proxies plus experimental PAC support, while plugins extend the available tools. Official releases are provided through GitHub, some platform packages are community-maintained, and nightly artifacts may be periodically removed.

LTX Desktop is an open-source desktop app for generating and editing videos with LTX models. It supports text-, image-, and audio-to-video generation, Retake editing, and project timelines. Windows or Linux systems with an NVIDIA CUDA GPU and at least 16 GB of VRAM can generate locally after downloading model weights; unsupported Windows hardware and macOS use the LTX API instead. The app is still in beta, local mode downloads large model files, and API-backed features send prompts and media to the service.