[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M72CDPCS0FTMRX1DN8TC5760":3,"skill-same:01M72CDPCS0FTMRX1DN8TC5760":46},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":26,"repositoryHtmlUrl":27,"repositoryHomepageUrl":28,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M72CDPCS0FTMRX1DN8TC5760","maude-positioning",null,"1aGh\u002Fmaude","1aGh","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F17166964?v=4","01M72CAQDBB0YG724KFT7WEYRF","Michal Dovrtel","01M72CAQDXC32AWQRZYNJXXDC4",3,"MIT","https:\u002F\u002Fgithub.com\u002F1aGh\u002Fmaude\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fmaude-positioning\u002FSKILL.md",[],[],"2026-11-10 18:39:20",false,"---\nname: maude-positioning\ndescription: Message architecture for talking ABOUT Maude the product (what it is, why it's great, how it differs from Figma \u002F Claude Design). Use whenever writing copy that pitches Maude itself: the homepage\u002Fdocs intro, a README top blurb, a changelog announcement, a social post introducing the project, an elevator pitch for a non-technical person, or any \"why is this different from Figma \u002F Claude\" explanation. Composes with michal-voice (that skill is the HOW\u002Ftone; this one is the WHAT\u002Fargument). Repo-internal, not shipped via the marketplace or npm.\n---\n\n# maude-positioning — the message architecture for pitching Maude\n\nRepo-internal skill (Maude-specific, lives in `.claude\u002Fskills\u002F`, not shipped via the marketplace or npm, same convention as `whats-new-entry` and `desktop-e2e`). This is the **WHAT** to say about Maude the product. Pair it with the personal `michal-voice` skill for the **HOW** (tone, tics, EN-dev mode) whenever the output is copy Michal is publishing himself.\n\n## When to use\n\nAny time you're about to write copy that pitches Maude itself, not a feature inside it: the homepage hero, a docs intro paragraph, the README top blurb, a changelog announcement post, a tweet\u002FLinkedIn post introducing the project, an elevator pitch for a non-technical friend, or anything answering \"what is this and why would I use it\" or \"how is this different from Figma \u002F Claude Design.\"\n\nNot for: internal command\u002Fskill descriptions, DDRs, code comments, or anything that isn't audience-facing marketing copy.\n\n## The core insight\n\nEvery other design tool treats AI as an addition. **Figma** is a human-first, cursor-and-mouse tool that bolted a copilot panel onto the side; the cloud file is still the source of truth. **Claude Design** (claude.ai\u002Fdesign and the wider chat-canvas family) is AI-first but ephemeral: a chat box rendering a preview, no git, no real files on disk, no persistent design system to check new work against, gone when the conversation scrolls past it.\n\nMaude starts from the other direction. It's a real, professional-grade tool (canvases are `.tsx` files, history is git, design tokens are CSS variables checked by critics) whose entire substrate was built for how an AI coding agent actually works: a dev-loop, a filesystem, a repo. Nothing here strains against the grain. That's the \"wet dream for designers\" feeling: the tool was built around the AI instead of the AI being bent to fit the tool.\n\n**Protect this one sentence through every rewrite:** built around AI, not bent to fit one.\n\n## Positioning statement\n\nFor someone already living inside Claude Code (a developer, or a designer pairing with one) who wants to design and ship UI without leaving the agent loop, Maude is a canvas-first design toolkit and agentic workflow system that treats the repo, not a proprietary cloud or a chat scroll, as the source of truth. Unlike Figma, the AI isn't a bolted-on feature. Unlike Claude Design, the work doesn't evaporate when the chat does.\n\n## Differentiation (cite only real, checkable facts)\n\n| Axis | Figma | Claude Design (claude.ai\u002Fdesign) | Maude |\n| --- | --- | --- | --- |\n| Where the AI sits | Bolted-on copilot panel | Is the whole product | Native substrate: every canvas, critic, and command is built for an agent loop |\n| Source of truth | Figma's proprietary cloud | The chat scroll (ephemeral) | Your git repo (`.tsx` canvases, `.ai\u002F` decisions, real commits) |\n| Persistence | Cloud file, vendor lock-in | None. Gone when the thread does | Git history, `\u002Fdesign:rollback`, snapshots per canvas |\n| Design system enforcement | Manual, or a paid plugin | None | Built-in critics (a11y, brand, tokens, motion...) gate every edit |\n| Team sync | Figma's hosted multiplayer (paid) | None | Optional **self-hosted** Yjs hub, no SaaS tier |\n| Beyond design | N\u002FA (design-only) | N\u002FA (chat-only) | `flow` runs the same repo-native loop for planning and shipping the whole feature, not just the mock |\n| Cost \u002F lock-in | Seat-based SaaS | Bundled into a Claude plan | Open-source, MIT, self-hostable, zero telemetry |\n\nKeep `flow` and the hub as the closing row of every version of this table: secondary to the design-tool hook, never dropped entirely.\n\n## Plain-language mode (non-technical audience)\n\nSkip \"canvas,\" \"dev server,\" \"critics.\" Use this shape instead: *\"Most design tools let you sketch a screen, then a developer has to rebuild it from scratch. AI chat tools let you describe a screen, but the second the conversation ends, so does the work. Maude designs the real screen with AI, and a developer can ship it as-is, because it was never a mockup to begin with. It's already the real thing.\"* Anchor on the rebuild-cost pain a non-technical stakeholder actually feels, not on technical plumbing.\n\n## Proof points (cite only shipped features, never a roadmap item)\n\n- Cmd+Click any element in a live canvas and edit it; it writes straight back to the source `.tsx`.\n- Full history\u002Frollback per canvas (`\u002Fdesign:rollback`), because it's git underneath.\n- A critic panel (a11y, brand, typography, motion...) grades every AI edit against the project's own design system before a human sees it.\n- Optional self-hosted sync hub for team multiplayer, Docker or Fly, no SaaS tier, no cloud middleman.\n- Open-source, MIT, zero telemetry, no signup.\n\n## Guardrails\n\n- **Accurate only.** Every claim above must map to a shipped feature. If some copy describes something not yet shipped, that's a bug in the copy, not a marketing choice: fix it or cut the claim.\n- **Confident, not sneering.** Contrast Figma and Claude Design on real, checkable differences (git, persistence, self-hosting). Never mock the products or their makers. \"Different tool, different job\" beats \"the other guy sucks.\"\n- **Never claim to replace Figma outright.** Maude doesn't do freeform vector editing or a from-scratch visual design practice. It's for the AI-agent-native slice: prototyping, iterating, and shipping UI from inside Claude Code.\n- **`flow` and the hub are always secondary, never absent.** The design-tool insight is the hook (the clearest wow factor). Every piece of copy should still surface `flow` and the self-hosted hub after the hook, not instead of it.\n- **Compose with `michal-voice`.** This skill decides the argument. `michal-voice` (EN-dev mode, for anything published in English) decides the sentence shapes: short declaratives, dry self-deprecation, **no em-dash as a clause joiner, ever**, no corporate gloss.\n","# maude-positioning — the message architecture for pitching Maude\n\nRepo-internal skill (Maude-specific, lives in `.claude\u002Fskills\u002F`, not shipped via the marketplace or npm, same convention as `whats-new-entry` and `desktop-e2e`). This is the **WHAT** to say about Maude the product. Pair it with the personal `michal-voice` skill for the **HOW** (tone, tics, EN-dev mode) whenever the output is copy Michal is publishing himself.\n\n## When to use\n\nAny time you're about to write copy that pitches Maude itself, not a feature inside it: the homepage hero, a docs intro paragraph, the README top blurb, a changelog announcement post, a tweet\u002FLinkedIn post introducing the project, an elevator pitch for a non-technical friend, or anything answering \"what is this and why would I use it\" or \"how is this different from Figma \u002F Claude Design.\"\n\nNot for: internal command\u002Fskill descriptions, DDRs, code comments, or anything that isn't audience-facing marketing copy.\n\n## The core insight\n\nEvery other design tool treats AI as an addition. **Figma** is a human-first, cursor-and-mouse tool that bolted a copilot panel onto the side; the cloud file is still the source of truth. **Claude Design** (claude.ai\u002Fdesign and the wider chat-canvas family) is AI-first but ephemeral: a chat box rendering a preview, no git, no real files on disk, no persistent design system to check new work against, gone when the conversation scrolls past it.\n\nMaude starts from the other direction. It's a real, professional-grade tool (canvases are `.tsx` files, history is git, design tokens are CSS variables checked by critics) whose entire substrate was built for how an AI coding agent actually works: a dev-loop, a filesystem, a repo. Nothing here strains against the grain. That's the \"wet dream for designers\" feeling: the tool was built around the AI instead of the AI being bent to fit the tool.\n\n**Protect this one sentence through every rewrite:** built around AI, not bent to fit one.\n\n## Positioning statement\n\nFor someone already living inside Claude Code (a developer, or a designer pairing with one) who wants to design and ship UI without leaving the agent loop, Maude is a canvas-first design toolkit and agentic workflow system that treats the repo, not a proprietary cloud or a chat scroll, as the source of truth. Unlike Figma, the AI isn't a bolted-on feature. Unlike Claude Design, the work doesn't evaporate when the chat does.\n\n## Differentiation (cite only real, checkable facts)\n\n| Axis | Figma | Claude Design (claude.ai\u002Fdesign) | Maude |\n| --- | --- | --- | --- |\n| Where the AI sits | Bolted-on copilot panel | Is the whole product | Native substrate: every canvas, critic, and command is built for an agent loop |\n| Source of truth | Figma's proprietary cloud | The chat scroll (ephemeral) | Your git repo (`.tsx` canvases, `.ai\u002F` decisions, real commits) |\n| Persistence | Cloud file, vendor lock-in | None. Gone when the thread does | Git history, `\u002Fdesign:rollback`, snapshots per canvas |\n| Design system enforcement | Manual, or a paid plugin | None | Built-in critics (a11y, brand, tokens, motion...) gate every edit |\n| Team sync | Figma's hosted multiplayer (paid) | None | Optional **self-hosted** Yjs hub, no SaaS tier |\n| Beyond design | N\u002FA (design-only) | N\u002FA (chat-only) | `flow` runs the same repo-native loop for planning and shipping the whole feature, not just the mock |\n| Cost \u002F lock-in | Seat-based SaaS | Bundled into a Claude plan | Open-source, MIT, self-hostable, zero telemetry |\n\nKeep `flow` and the hub as the closing row of every version of this table: secondary to the design-tool hook, never dropped entirely.\n\n## Plain-language mode (non-technical audience)\n\nSkip \"canvas,\" \"dev server,\" \"critics.\" Use this shape instead: *\"Most design tools let you sketch a screen, then a developer has to rebuild it from scratch. AI chat tools let you describe a screen, but the second the conversation ends, so does the work. Maude designs the real screen with AI, and a developer can ship it as-is, because it was never a mockup to begin with. It's already the real thing.\"* Anchor on the rebuild-cost pain a non-technical stakeholder actually feels, not on technical plumbing.\n\n## Proof points (cite only shipped features, never a roadmap item)\n\n- Cmd+Click any element in a live canvas and edit it; it writes straight back to the source `.tsx`.\n- Full history\u002Frollback per canvas (`\u002Fdesign:rollback`), because it's git underneath.\n- A critic panel (a11y, brand, typography, motion...) grades every AI edit against the project's own design system before a human sees it.\n- Optional self-hosted sync hub for team multiplayer, Docker or Fly, no SaaS tier, no cloud middleman.\n- Open-source, MIT, zero telemetry, no signup.\n\n## Guardrails\n\n- **Accurate only.** Every claim above must map to a shipped feature. If some copy describes something not yet shipped, that's a bug in the copy, not a marketing choice: fix it or cut the claim.\n- **Confident, not sneering.** Contrast Figma and Claude Design on real, checkable differences (git, persistence, self-hosting). Never mock the products or their makers. \"Different tool, different job\" beats \"the other guy sucks.\"\n- **Never claim to replace Figma outright.** Maude doesn't do freeform vector editing or a from-scratch visual design practice. It's for the AI-agent-native slice: prototyping, iterating, and shipping UI from inside Claude Code.\n- **`flow` and the hub are always secondary, never absent.** The design-tool insight is the hook (the clearest wow factor). Every piece of copy should still surface `flow` and the self-hosted hub after the hook, not instead of it.\n- **Compose with `michal-voice`.** This skill decides the argument. `michal-voice` (EN-dev mode, for anything published in English) decides the sentence shapes: short declaratives, dry self-deprecation, **no em-dash as a clause joiner, ever**, no corporate gloss.\n",{},"main","cfde9d2b169e4b418a5c5efb35297cc575dc764c",".claude\u002Fskills\u002Fmaude-positioning",0,"https:\u002F\u002Fgithub.com\u002F1aGh\u002Fmaude","https:\u002F\u002Fmaude.sh","https:\u002F\u002Fgithub.com\u002F1aGh","www.iagh.cz","2026-11-13 20:38:15",{"source":7,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add 1aGh\u002Fmaude --skill maude-positioning",{"tool":38,"command":39},"bunx","bunx skills add 1aGh\u002Fmaude --skill maude-positioning",{"tool":41,"command":42},"pnpm","pnpm dlx skills add 1aGh\u002Fmaude --skill maude-positioning","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002F1aGh\u002Fmaude\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fmaude-positioning\u002FSKILL.md",true,[],[47,54,61,68,75,82],{"skillNo":48,"name":49,"nameZh":6,"description":50,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":51,"categories":52,"keywords":53,"publishedAt":18},"01M72CJCVNC5JB3M36ES649CTY","workflow-state","Read and update workflow state when planning, executing, completing, pausing or resuming project work.","https:\u002F\u002Fgithub.com\u002F1aGh\u002Fmaude\u002Fblob\u002Fmain\u002Fplugins\u002Fflow\u002Fskills\u002Fworkflow-state\u002FSKILL.md",[],[],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":18},"01M72CJAJBZT2RHE97ND8EFTA3","ui-rules","UI hard-stop rules and audit method: WCAG 2.1 AA accessibility, responsive layout and motion.","https:\u002F\u002Fgithub.com\u002F1aGh\u002Fmaude\u002Fblob\u002Fmain\u002Fplugins\u002Fflow\u002Fskills\u002Fui-rules\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":18},"01M72CJ857314HCC8A99T4K0D2","testing-rules","Apply test quality rules during implementation and review: meaningful assertions, isolation and coverage.","https:\u002F\u002Fgithub.com\u002F1aGh\u002Fmaude\u002Fblob\u002Fmain\u002Fplugins\u002Fflow\u002Fskills\u002Ftesting-rules\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":18},"01M72CJ6N80HW2WRS8BV4QSS4V","security-rules","Apply application and AI security checks to code reviews, implementation and threat-driven validation.","https:\u002F\u002Fgithub.com\u002F1aGh\u002Fmaude\u002Fblob\u002Fmain\u002Fplugins\u002Fflow\u002Fskills\u002Fsecurity-rules\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":18},"01M72CJ4C41Q4E1WDTFMKYFX3Q","scenario","Author or run cross-platform UI scenarios with selectors, screenshots, platform coverage and result reports.","https:\u002F\u002Fgithub.com\u002F1aGh\u002Fmaude\u002Fblob\u002Fmain\u002Fplugins\u002Fflow\u002Fskills\u002Fscenario\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":18},"01M72CJ1QYX752BM7G12250YJR","question-protocol","Batch necessary questions from workflow participants, collect user answers and route them to the requester.","https:\u002F\u002Fgithub.com\u002F1aGh\u002Fmaude\u002Fblob\u002Fmain\u002Fplugins\u002Fflow\u002Fskills\u002Fquestion-protocol\u002FSKILL.md",[],[]]