Design Tools

Excalidraw135.2k★

Excalidraw is a virtual whiteboard for sketching diagrams, wireframes and flowcharts in a hand-drawn style. It runs in your browser, it's fast, and the sketchy look makes diagrams read as intentionally informal instead of accidentally ugly. A community library of importable shape packs (AWS icons, network diagrams, UI kits) covers the common cases. MIT licensed. The free editor at excalidraw.com gives you the full drawing tool, real-time collaboration without an account, and export to PNG, SVG or the native .excalidraw format. There's an npm package to embed it in your own app. Self-hosting is a static site, but a default build still points share links, live collaboration and AI at Excalidraw-run servers; a private install means repointing those and running your own collaboration server. Excalidraw+ is $7 per user a month, or $6 billed annually, and buys unlimited saved scenes and folders, cloud storage, access control, edit and view sharing, comments, presentations and team management. Shared workspace libraries are still marked coming soon. Solo and small teams: the free editor plus exported files is plenty. Growing teams that want everything saved in one place: pay. Large: check whether Figma or Miro already covers this. The catch: it's a drawing tool, not a diagram engine. It has elbow arrows, flowchart shortcuts and Mermaid paste, but there's no auto-layout and nothing updates from code or data. For diagrams as code, look at Mermaid or D2.

open-design106.2k★

Open Design is a local-first tool that turns your AI coding agent into a design studio. Point it at Claude Code, Cursor, Codex, or any of twenty-plus agents it supports, and it generates UI prototypes, dashboards, slide decks, images, and even video, all on your own machine, with no proprietary cloud in the middle. It's Apache 2.0 and positions itself as the open answer to tools like Vercel's v0 and Figma Make. The recent work has pushed it well past a simple prototype generator. It now ships with 150-plus built-in design systems, so outputs can auto-adopt a specific brand's tokens and look, plus a library of prebuilt skills for common jobs and a growing plugin marketplace. Video comes through HeyGen. Under the hood it's a Next.js and Electron app you can run as a desktop install or deploy with Docker. Setup is moderate, Node and pnpm, more moving parts than a single CLI but not painful. You bring your own model keys, and the media features lean on outside services like HeyGen that carry their own costs. It's genuinely useful if you want AI-generated design work without handing your prompts and outputs to a hosted platform. Solo builders and small teams experimenting with agent-driven design are the sweet spot. The catch is maturity. This is still a 0.x project moving fast, with a broadening scope and a healthy pile of open issues. The brand-design-system layer is brand new. If you need something stable and supported today, v0, Framer AI, or Lovable will feel more finished, at a subscription cost. If you want local-first and open, and you can tolerate the rough edges, Open Design is one of the few real options.

photocraft77.3k★

PhotoCraft is a Photoshop lookalike written from scratch in Rust: same menus, same shortcuts, layers, masks, adjustment layers and layer styles, and it opens and saves real PSD files. It runs on Mac, Windows, Linux, FreeBSD and in the browser, free under MIT or Apache 2.0, your pick. There's nothing to host. Download an installer and go; Mac and Windows builds are now signed. It also reads Krita, Paint.NET and Affinity files, loads Photoshop actions (.atn), and ships a command line tool and an MCP server, so an AI agent can drive the same commands the menus use. Hobbyists who just need to open a PSD without paying Adobe should try it now. Professionals shouldn't switch yet; the team's own roadmap rates it about 45% ready for real work, with about a dozen tools still missing, no generative fill and no .8BF plug-in support. GIMP and Krita are the mature free options. The catch: it launched in October 2026 and users file bug reports by the hundreds. Much of the code was written with Claude, which has people questioning the "clean-room" label. It's also the free front door to ArtCraft, the team's AI image and video studio, where subscriptions are optional, and nothing here is gated.

Penpot62.1k★

Penpot is the open source Figma alternative: browser-based design and prototyping with components, design tokens, CSS Grid and Flex layouts, and real-time collaboration. Designs use open standards (SVG, CSS, HTML), so developers inspect real code instead of guessing. The core is MPL-2.0, and both the cloud plan and self-hosting are free. Self-hosting is a Docker Compose stack: frontend, backend, exporter, Postgres and Valkey. Penpot says to plan 50 to 100GB of storage for up to 10 editors. Upgrades are on you, and here they matter. The free cloud plan caps a team at 8 members (viewers unlimited) and 10GB of storage. Unlimited is $7 per editor a month, capped at $175. SSO and the admin console need Enterprise at $25 per member a month. Solo and small teams: free. Medium teams: self-host or pay for Unlimited, well under Figma's $16 Full seat. Large orgs: Enterprise. The catch: Penpot published 24 security advisories in 2026, including an account takeover any registered user could pull off and a 9.9-rated command injection in the exporter. All are fixed in 2.18, but a forgotten self-hosted instance is exposed. Figma's plugin library is still far deeper.

tldraw51.3k★

tldraw is the React SDK behind the infinite canvases in products like ClickUp, Replit and Google Stitch. Drop in one component and you get shapes, arrows, freehand drawing, rich text, image export and real-time multiplayer, then swap almost any of it for your own shapes, tools and UI. It is free to build with. It is not open source. Setup is npm i tldraw and a <Tldraw /> component. Multiplayer runs on tldraw sync, which you host yourself; the starter kit uses Cloudflare Durable Objects, the same stack as tldraw.com. Production is where the license bites. Every production deployment needs a key: a free hobby key for side projects, research and prototypes (reviewed application, "made with tldraw" watermark), a 100-day free trial, or a commercial license with unpublished, value-based pricing and startup discounts. Solo hobbyists: free with the watermark. Small teams and larger teams shipping a product: budget for a commercial license before launch. The catch is the license history. tldraw was Apache 2.0 until December 2023, then free for commercial use with a watermark, and since 4.0 in September 2025 production needs a key, enforced in code. Excalidraw is MIT with no key.

GrapesJS26.4k★

GrapesJS is the drag-and-drop editor you put inside your own product, not a website builder you sign up for. It ships the canvas, block manager, style manager, layer tree, code viewer, and asset manager as a BSD-3 licensed JavaScript library. Free, no key, no session cap, and shipping since 2016. You supply everything around it. Storage, publishing, hosting, templates, user accounts, and asset uploads are yours to build, and the plugin ecosystem is large precisely because the core stops at the editor. Budget weeks, not days, to get from a working init call to something a customer would touch. Solo devs and small teams building a builder get a lot for nothing. Bigger teams should price the alternative first: the same maintainers sell Studio SDK, a finished embeddable editor with support attached, whose free tier stops at 1,000 sessions a month with Studio branding. Startup is $200/month and Business is $2,000/month. The catch is what this is not. GrapesJS does not replace Webflow for anyone who just wants a website: no hosting, no CMS, no publish button. What it replaces is the six months you would spend writing a visual editor from scratch. webstudio-is/webstudio is where to look for the finished product.

Compositor18.5k★

Compositor is a free, native Mac image editor built to feel like Photoshop, down to the keyboard shortcuts, so the muscle memory carries over. Layers, folders, masks, clipping masks, adjustment layers for curves and levels, a clone stamp, spot healing, content-aware fill, subject selection, and background removal all work today. MIT licensed, no account, no subscription. Install is a signed and notarized DMG from GitHub Releases with automatic updates. The hard requirement is macOS 26.5. Anything older will not run it, and an open issue asks for macOS 15 support. It is Mac only. It imports JPEG, PNG, HEIC, TIFF, and as of version 1.2 Photoshop PSD files, keeping folders, masks, and some blend modes editable. Solo creators and small teams who composite and retouch photos: try it before renewing Adobe. Anyone who trades PSDs with clients should wait. On Windows or Linux, GIMP and Krita are the established open source options, and penpot/penpot covers UI design, not photos. The catch: it was five days old when we looked, shipping several releases a day. PSD support is import only, 8-bit RGB only, no CMYK, and PSD text layers turn into pixels. There is no PSD export at all. The 1.2 release notes admit layer effects were being lost on save until that fix. Keep backups of anything that matters.

drawnix15.0k★

Drawnix is a free whiteboard in your browser for mind maps, flowcharts and freehand sketches. Nothing to sign up for: open drawnix.com and start drawing, and everything saves in your browser. It can also turn Mermaid code into a flowchart and a Markdown outline into a mind map, which is the feature to try first. MIT, backed by PingCode, the company behind the Plait drawing framework it's built on. Self-hosting is a Docker image or a static build, so there's no backend to babysit. Exports cover PNG, JPG, SVG and its own .drawnix file. The toolbar has shapes, straight, curved and elbow arrows, text, pen, eraser and images, plus six color themes. Solo users who think in mind maps: it fits better than excalidraw/excalidraw, which is freeform first. Teams that need to draw together live should stick with Excalidraw or tldraw/tldraw, because Drawnix has no real-time collaboration. The catch is pace and storage. Releases are slow, seven months between the last two, and browser storage means clearing your site data can wipe your boards. Export anything you care about.

Kami12.3k★

Kami is a design system for professional documents. You feed it content through Claude Code or another AI agent, and it outputs HTML in a warm-parchment, ink-blue style. It now ships nine template types: one-pagers, long documents, formal letters, portfolios, resumes, slides, equity reports, changelogs, and landing pages. MIT-licensed. Installation is loading one skill into your agent of choice, available for Claude Code, Claude Desktop, Codex, and generic agent frameworks. No app to run. The system enforces typography and spacing rules so output stays consistent whether you are writing a one-page memo or a long report, and it now includes built-in SVG diagram types plus Mermaid support for sequence, class, and ER diagrams. HTML output means you print directly or export to PDF. English, Chinese, and Korean are all supported. Writers and analysts who need polished docs without opening InDesign get a sensible default. Job seekers get resume templates that do not look like every other Canva export. Solo creators get a style system they can extend. No paid tier. The catch: the Chinese serif font (TsangerJinKai02) is free for personal use only. Commercial use of that font requires licensing from tsanger.cn. Documents in English or Korean, and Chinese docs using a different font, are fully unrestricted.

vectorcraft11.4k★

VectorCraft is a free, from-scratch rebuild of the Adobe Illustrator workflow: the Pen tool, Pathfinder, Appearance panel, live blends, type, and SVG, PDF and .ai import and export, in a native Rust app for macOS, Windows, Linux and FreeBSD, plus a browser build. It's dual-licensed MIT or Apache 2.0, from the ArtCraft team behind PhotoCraft. Setup is a download. Signed installers, AppImage, Flatpak, .deb and .rpm sit on the GitHub releases page, or build it with cargo run --release -p vectorcraft. It also ships an MCP server, so Claude and other agents can draw, edit and export like a person does. There's no paid tier. ArtCraft sells AI generation credits for a separate app, ArtCraft Studio. Solo illustrators and small teams: free, but test it on throwaway work first. Larger teams living in Illustrator: not yet. The catch: it's alpha and about a month old. The README rates it roughly two-thirds ready for real work, Illustrator's own .ai files can open with the wrong layer structure and can't be saved back with Illustrator's editing data, and some machines fail to start the app. Inkscape is still the safer free pick for client work.

Krita10.7k★

Krita is a digital painting application built for artists rather than photographers, and that focus is the whole point. Over a hundred brush engines, stabilizers for shaky hands, wrap-around mode for seamless textures, a proper animation timeline with onion skinning, and HDR painting. It reads and writes PSD. GPL-3.0, cross-platform, and the current stable release is 5.3.4. Installation is a download on Windows, macOS, and Linux, with an Android tablet build as well. Nothing to configure and nothing to run. It expects a drawing tablet: pressure sensitivity is where the brush engines earn their reputation, and using it with a mouse wastes most of what makes it good. GitHub is a read-only mirror: bugs and patches go to KDE's Invent instance, and pull requests opened here get closed unmerged. Illustrators, concept artists, and animators should use this instead of Photoshop or Clip Studio Paint. It is not a photo editor, so retouching and compositing belong in GNOME/gimp. Studios running Windows can put it on every machine at no cost and no seat counting. The catch is where the money goes, and it is worth understanding. Krita is free from krita.org, but paid builds also sit on Steam, the Microsoft Store, Epic, and the Mac App Store. Those are the same GPL program; you are buying a donation with automatic updates attached. Paying is a choice to fund development, never a way to unlock anything.

m3e-canvas10.2k★

M3E Canvas is a browser sketchpad for Material 3 Expressive screens (Google's latest design language) that turns your drawing into a prompt for an AI coding tool. You drag buttons, bars, and cards onto a phone or desktop frame, link them into a tappable flow, then copy a plain-language brief you paste into Claude Code, Codex, or Cursor to build the real app. It is MIT and runs entirely in your browser. The clever part is the handoff. Instead of exporting design files no AI can read, it writes a concise natural-language description of the whole flow (layouts, navigation, theme) in English, Japanese, or Chinese, so the coding tool has real instructions. There is no backend; everything saves to localStorage. An optional AI helper writes behavior notes if you bring your own model key. If you vibe-code UIs and keep getting layouts that ignore your intent, this is a fast way to pin down what you actually want before the model starts. Solo builders and small teams get the most out of it. The catch: it only speaks Material 3 Expressive. If you are not building an Android-flavored UI, the output fights whatever design system you are actually using.

infinite-canvas9.8k★

Infinite Canvas is a freeform AI creation board: drop text, image, video and audio generation nodes onto an endless canvas, edit with reference images, and keep a prompt library and asset shelf alongside. You bring an OpenAI-compatible endpoint for the actual generation. It switched from AGPL to MIT in August 2026, so commercial and closed-source use is now fine. It's now a frontend-only app: deploy it to Vercel or run it in Docker, and projects, assets and your API key live in your browser. There's a public demo at canvas.best. Newer additions include a Canvas Agent that lets Codex or Claude Code operate the canvas over MCP, a plugin system and WebDAV sync. Free, with paid custom development available from the maintainer. Solo creators who already pay for an image or video API get a flexible free front end. Teams should wait: the maintainers say data compatibility between versions isn't guaranteed and recommend forking if you need stability. The catch: the project openly courts terms-violating backends. Its description advertises chatgpt2api and grok2api channels, and the README points to the owner's own chatgpt2api, a reverse-engineered ChatGPT web proxy that warns users about account bans. Point it at an official API and it's a fine tool; point it at those and you're the one taking the risk.

Webstudio9.2k★

Webstudio is an open-source visual website builder, and it says the quiet part out loud in its own tagline: it's a Webflow alternative. Drag-and-drop canvas, control over every CSS property, connect any headless CMS, and here's the kicker, it exports every project as a standard Remix/React app you can host anywhere. AGPL, free, no lock-in on what you build. Running it yourself is real work. This isn't a static template, it's a full TypeScript app with a builder, a database, and asset storage. Most people use the hosted version, which has a free Hobby tier (custom domains, a few thousand pageviews a month) and a Pro tier around $15/mo that unlocks CMS integrations, client editing mode, and staging. Self-hosting is possible and truly lock-in-free, but budget the ops time. For designers and small teams who want Webflow's visual power without Webflow's bill, or who need to own their hosting, this is the strongest open-source option going. Solo and small teams do fine on the free Hobby tier or cheap Pro. Larger teams weighing self-host should price the ops time against the hosted seats. Two catches. The AGPL license is copyleft, so if you self-host and modify it inside a commercial product, read the terms first. And the most convenient path, hosted with custom domains and CMS, is the paid one, so "free" really means "free if you do the hosting yourself." The builder is excellent; the freedom has an ops price tag.

open-codesign8.2k★

Open CoDesign turns text prompts into design artifacts on your desktop. Prompt to React prototype, slide deck, PDF, or marketing asset, all rendered in a sandboxed iframe with a live preview pane. It pitches itself as the open-source Claude Design alternative, MIT-licensed and built around your own model API key. Install in under 90 seconds via Homebrew, winget, or direct download. First launch asks for an API key (Claude, OpenAI, or 20+ providers), paste and go. Everything runs locally; credentials live in ~/.config/open-codesign/config.toml and nothing leaves your machine unless your model provider needs it. Fifteen built-in demos cover landing pages, dashboards, and mobile UIs to get you started. Solo designers and small teams get a real Claude Design substitute without the subscription. You pay model tokens only. Larger teams that need brand consistency or shared component libraries should still look at Figma plus AI plugins, since this is built around individual workflows. The catch: it's young and the Electron app is heavier than a web tool. If you want a polished web design IDE, Claude Design is more refined, but it's $20+ a month and locks you to Anthropic models.

cadcraft7.7k★

CADCraft is a from-scratch take on AutoCAD's 2D drafting workflow, written in Rust. You get the command line (type L, click two points), object snaps, layers, dimensions, hatches, blocks and paper-space layouts that plot to PDF. It's free under MIT or Apache-2.0, with builds for macOS, Windows, Linux, FreeBSD and the browser. Setup is a download from the releases page, or a Cargo build from source. A CLI and an MCP server let an AI agent type commands and render the drawing to check its own work. Fitting, since AI agents wrote most of it: the roadmap says the first 51,000 lines took about 11.5 agent-hours. There's no paid tier. Solo hobbyists and students sketching 2D plans: free, worth a look. Small firms and larger teams exchanging drawings with clients: keep AutoCAD, or try LibreCAD if you need something free and stable today. The catch: the project calls itself pre-alpha and puts itself at about 38% ready for real work. DWG is the format that matters, and a DWG saved here drops everything CADCraft doesn't model and has never been checked in AutoCAD. Open a file a client sent, sure. Don't send one back.

plasmic7.2k★

Plasmic is a visual builder for React apps and sites that plugs into your real codebase. Designers and non-developers drag and drop UI, developers wire in custom React components and data, and nobody's stuck exporting dead mockups. The client libraries are MIT and free; the Studio you actually build in is a hosted app. Self-hosting is technically on the table, the platform directory is AGPL, but it's a full app plus backend plus database and it's barely documented for self-host. In practice you use the hosted Studio. So read this less as "free self-hosted tool" and more as "SaaS with an open client." The free tier is real: unlimited projects, a headless CMS, custom domains, three collaborators, with a Plasmic badge on your site. Paid starts at $39/mo to drop the badge, $103/mo for a bigger team and AI, and $399/mo for A/B testing and scheduling. The gates are seats, advanced content features, and branding. Solo builders and tiny teams do fine free. The moment you need a fourth collaborator or want the badge gone, you're paying. The catch is the one every open-core visual builder has: the free tier is a real product right up until your team grows, and then the seat cap does exactly what it's designed to do. If you want an open alternative to Webflow or Builder.io that lives in your codebase, Plasmic is a strong one. Just go in knowing the hosted Studio is the actual product.

GIMP6.6k★

GIMP is the free image editor people reach for when they will not pay Adobe. Photo retouching, compositing, masks, layers, curves, batch processing through a scripting console, and a plugin ecosystem going back decades. Version 3.2 is current, and it finally brought non-destructive editing, better HiDPI handling, and a GTK3 interface that no longer looks like 2008. GPL-3.0, genuinely free, no account and no tier. The catch most people hit is not capability, it is muscle memory. GIMP does not copy Photoshop's shortcuts or its single-window habits, so switching costs you a week of fumbling before your hands know where things are. CMYK support is still weak, which rules it out for most print work. Running it is trivial: packages exist for every desktop, plus a Flatpak. Solo creators and small teams doing web graphics, screenshots, and photo cleanup should stop paying for Photoshop and use this. Anyone doing print production or trading layered PSDs with clients will keep fighting it. For digital painting specifically, KDE/krita is the better tool. The catch worth knowing before you file a bug: GitHub is a read-only mirror. Development, issues, and merge requests all live on GNOME's GitLab, so the activity signals here understate the project by a wide margin.

design-extract4.8k★

Reverse-engineers any website to extract its complete design system. Feed it a URL and it pulls out color tokens, typography scales, spacing values, component patterns, and Tailwind configs. Exports to Figma variables, React themes, or raw JSON. One npm install, one command. Runs headless Chrome to capture the site across viewports and analyzes the computed styles. Includes accessibility scoring for contrast ratios. Solo developers and designers who want to study how production sites are built will get the most out of this. Zero cost, MIT licensed. The catch: it captures what's rendered, not what was intended. Computed styles can be messy, and auto-generated Tailwind configs may need cleanup before use in production.

figcraft4.5k★

FigCraft is a from-scratch Figma lookalike written in Rust: frames, auto layout, components and variants, vector networks, styles, variables and clickable prototypes, in a native, offline app. It's free under MIT or Apache-2.0, with no account and no telemetry, and your files are plain .figcraft archives on your own disk. Grab a release for any desktop OS, or run it in the browser. Real-time collaboration is a preview: share through the default public server or run your own figcraft-server. There are no accounts, so whoever has the link has access, and the server stores files unencrypted. A CLI and MCP server expose 400+ commands to AI agents. No paid tier. Solo designers and developers who want offline mockups: free, try it. Small teams living in Figma multiplayer: not yet; Penpot is the more mature open source pick. Larger teams: no accounts, so no SSO or permissions. The catch: Figma's .fig files open best effort only, and the project says that path hasn't been checked against real-world files. The maintainers rate it about half ready for real work, and its first version was built by parallel AI agents in about 25 hours. Test your own files before trusting it.

doory2.4k★

Doory is a Canva-style design editor you can run yourself: posters, social posts, slides, resumes, business cards and short animations, exported as PNG, PDF, MP4 or GIF. There's no paid plan, no watermark, and no account needed to start. A free hosted copy runs at doory.art. Self-hosting is a Docker Compose stack: a Django API, a real-time server for live co-editing, Postgres and nginx. Set two secrets, bring it up, create an admin. Then seed it, because a new site starts empty: fonts, icons, the background-removal model and templates are each separate download commands. Brand kits, share links, comments and version history all come free. Solo: use the hosted site. Small teams: self-host to keep designs on your server. Larger teams: not yet. Sign-in is email or Google, no SAML or OIDC, and ten starter templates is a long way from Canva. The catch is the license. It's the Attribution Assurance License, not MIT, and the author reads it as requiring a visible "Doory by Siam" credit in the running app for anyone who hosts it for users. Fine internally, a dealbreaker for white-labeling. The NOTICE file even calls the code MIT, contradicting the LICENSE. And it's five days old.

cad-viewer1.2k★

cad-viewer opens AutoCAD drawings (DXF and DWG files) in a web browser and lets you edit them, with no server. The file is parsed on the user's own device and never uploaded. It ships as npm packages and a one-line iframe embed, so you can drop a CAD viewer into your own app. MIT. DXF works out of the box with the project's own MIT parser. DWG, the format most people actually have, is where licensing kicks in: the open source route uses LibreDWG, which is GPL-3.0, adds roughly 13MB to the download, and still covers drawing entities only partially. There's also an AI assistant plugin that draws lines, circles and layers from a chat panel, using your own OpenAI or Anthropic key. Internal tools and open source projects: use it freely, DWG included. Closed-source products that need DWG either comply with the GPL or buy the maintainer's commercial parser: $3,000 for a perpetual license, then $1,500 a year if you want upgrades. The catch: very large DWG files can run the browser out of memory, and the commercial parser's trial licenses go only to companies, not individuals.

sticker-forge754★

Sticker-forge makes die-cut stickers you can peel off the screen, complete with a satin backing, realistic lighting, and a little peel sound. Build them from text or uploaded images in a WebGL canvas, and because it ships a framework-agnostic web component and JS API, you can embed the whole effect in your own site rather than just using the hosted maker. The engineering is nicer than the premise suggests: custom WebGL shaders for the peel physics, HEIC image conversion, and synthesized audio for the peel itself. MIT licensed, and the hosted demo is free to play with. Solo or team, it costs nothing. Use it when you want a playful, tactile touch, a fun sticker generator on a marketing page, or interactive stickers in a product. The catch is that this is a novelty, and it doesn't pretend otherwise. It's one very well-executed WebGL effect. Delightful, genuinely fun to use, but not a tool that earns a place in a serious workflow. Grab it when you want a moment of polish, not a feature.

Glaxnimate43★

Glaxnimate is a free desktop editor for making Lottie animations, the lightweight vector animations that play in apps and on the web. LottieFiles is the usual on-ramp here, but it's a hosted platform that meters exports and nudges you toward paid plans. Glaxnimate runs on your own machine, exports the same Lottie JSON, and costs nothing. GPL licensed, now maintained under the KDE project. There's no server and no account: you install it and open it like any other app on Windows, Linux, or macOS. It edits vector shapes on a timeline and exports to Lottie, SVG, animated PNG, GIF, or video. It won't feel as polished as a commercial tool, the interface is functional rather than slick, and serious motion-graphics work still belongs in After Effects with the Bodymovin exporter. For icons, loaders, and simple UI animation, it does the job without a subscription. This is the right tool if you make the occasional Lottie animation and don't want a hosted platform metering your exports or watching over your shoulder. Solo developers and small teams: this covers you for free. Larger design teams already living in After Effects will keep using that plus Bodymovin, and reach for Glaxnimate when they want a quick standalone edit without firing up the Adobe stack. The catch is reach. Glaxnimate's real home is GitLab under KDE, so its GitHub presence looks tiny and the project moves at the pace of a small volunteer team. It's stable and actively maintained, but don't expect the tutorial library, the marketplace, or the collaboration features a hosted platform sells you. You're trading the ecosystem for ownership and a price of zero.