[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CSDVB2569GJ65KYF7VAHNT":3,"skill-same:01M7CSDVB2569GJ65KYF7VAHNT":48},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":30,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M7CSDVB2569GJ65KYF7VAHNT","design-conformance",null,"Use when your context carries \"The approved design this task builds\" - comparing the running UI with the approved hand-off and mockup at each width and state, evidence screenshots attached to the task, Blocker\u002FHigh\u002FMedium\u002FNitpick findings and which of them fail the round","makifbaysal\u002Ftasktrooper","makifbaysal","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F25031940?v=4","01M7CRG3ZCSYZSSEVDMW3DEYC8","Mehmet Akif Baysal","01M7CRG401FJ8YAXH5ER730XYV",122,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fmakifbaysal\u002Ftasktrooper\u002Fblob\u002Fmain\u002Fcatalog\u002Fagents\u002Fqa-agent\u002Fskills\u002Fdesign-conformance\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: design-conformance\ncategory: qa\ndescription: Use when your context carries \"The approved design this task builds\" - comparing the running UI with the approved hand-off and mockup at each width and state, evidence screenshots attached to the task, Blocker\u002FHigh\u002FMedium\u002FNitpick findings and which of them fail the round\n---\n# Design Conformance\n\n## Overview\n\nA task built from an approved design is tested against that design too. A human approved it before any code was written, and its hand-off spec is the contract the developer built to — an oracle like the acceptance criteria and the analysis spec. You read the design documents, never the code (black-box-testing), and compare what the running product shows with what they say.\n\n**Core principle:** The approved design is the expected result. Same widths, same states, side by side — every finding cites the hand-off line or the mockup frame it contradicts, and the screenshot that shows it.\n\n## Inputs\n\n- **The hand-off spec** — `handoff: \u003Cscreen>`, first under \"## The approved design this task builds\": the components, the tokens, every state, the breakpoints, the exact copy. Its header names the chosen mockup, `design: \u003Cscreen> · \u003Cletter>`; the other variant documents were not chosen — never test against them.\n- **The chosen mockup** — its text rendition follows the hand-off in your context. To see a frame's look, render it under `$QA`, never the workspace: `list_task_documents \u003Cdesign key>` with `raw: true` (follow `next_offset` to the end), write it with a quoted heredoc (`mkdir -p \"$QA\u002Fdesign\"; cat > \"$QA\u002Fdesign\u002F\u003Cslug>.html\" \u003C\u003C'TT_HTML'` … `TT_HTML`), serve it on loopback (`nohup python3 -u -m http.server 0 --bind 127.0.0.1 --directory \"$QA\u002Fdesign\" > \"$QA\u002Fdesign.log\" 2>&1 & echo $! > \"$QA\u002Fdesign.pid\"`, the port is in the log), `browser_navigate` there and screenshot the frame you compare. Stop it by its PID (stop-what-you-started).\n- **The task's scope** — where the task description and the design disagree, the description is the narrower scope and wins for this task; a part of the design outside it is not a finding.\n\n## Cases\n\nAdd them to the matrix before you boot anything (scenario-plan-first): category `visual`, title prefix `design:` (`design: invoices empty 375 -> matches #a-375-empty`), linked to the criterion that names the state or the copy when there is one — the designer's hand-off criteria usually do. One case per state the hand-off lists, per width:\n\n- **Web:** 1440 (`browser_set_viewport {device:\"desktop\"}`), 768 (`{device:\"tablet\"}`), 375 (`{device:\"mobile\", width:375}`) — the design's widths. The four-width overflow check (360 · 768 · 1024 · 1440, ui-visual-evidence) still runs on its own.\n- **Mobile:** the phone frames on the device — portrait, and landscape where the hand-off draws it — and the tablet frames when the device or simulator is a tablet. A width you cannot reach is `skipped` with the reason, never assumed to match.\n- **States:** default, loading, empty, error, success, dark, the longest content — each one the hand-off lists, reached from the outside: an empty search, a filter with no result, an invalid submit, a failing dependency (frontend-manual-testing's `route` recipe). A state you cannot reach is `skipped`, \"not reachable from outside: \u003Cwhy>\".\n\n## Compare, in this order\n\n1. **Layout and hierarchy** — the regions, their order, what comes first, the one primary action.\n2. **Copy** — every string verbatim against the hand-off's Copy table; a paraphrase is a deviation.\n3. **Components** — the one the hand-off names, in its variant (`browser_read_dom` \u002F `mobile_read_ui` shows its role and label).\n4. **Tokens** — colour, type size, spacing, radius as the design system holds them (`get_design_system`); `browser_read_dom` with `as_text: false` shows the classes and styles an element carries. A colour the design system does not hold is a deviation.\n5. **States** — every listed state present, and drawn as designed.\n6. **Responsive behaviour** — per breakpoint the hand-off lists.\n7. **Visible accessibility it names** — labels, focus ring, status never colour alone, touch targets.\n\n## Evidence\n\nEvery comparison shot of the running UI goes on the task, where the human sees it: `browser_screenshot` (`full_page: true`, no `width` after `browser_set_viewport`) or `mobile_screenshot`, with `attach_to_task: true` and `title: \"\u003Cscreen>-\u003Cstate>-\u003Cwidth>\"` (`invoices-empty-375`). Cite it by that title in the case's `evidence` and the criterion note, with the URL, the viewport report and what it showed — a title, never a path.\n\n## Severity — what fails the round\n\n| Severity | Means | Example | Verdict |\n|---|---|---|---|\n| Blocker | the screen fails the approved design on a core path | the primary action missing at 375; the layout breaks at 768; body text fails contrast | case `failed`, criterion rejected, need_revision |\n| High | a deviation users will notice, or a design-system violation | the wrong component; an off-token colour; a missing empty state; key copy wrong | case `failed`, criterion rejected, need_revision |\n| Medium | noticeable, minor | spacing one step off; secondary copy paraphrased; alignment | case `passed`, the deviation in its `notes` and the criterion note |\n| Nitpick | polish | icon optical alignment; a 1px border difference | same as Medium |\n\nA Blocker or High fails the round like any other failure (bug-report-writing, qa-fail-to-need-revision): the case `failed` with `actual` = what the screen shows, the criterion it breaks rejected with expected (the hand-off line, or the frame id such as `#a-375-empty`) vs actual and the screenshot title, and one line in the numbered need_revision index — `3. [AC4 · case \"design: invoices empty 375\"] empty state has no \"New invoice\" action — High`. A failed design case with no criterion to reject still fails the round. Medium and Nitpick never fail the task: they are notes, like any polish beyond the floor.\n\nWrite each finding as a problem, not a prescription: what the design says, what the screen shows, where (URL + width + region) — never \"change the class\".\n\n## Red Flags\n\n- Testing against a variant document that was not chosen.\n- A `design:` case passed with no attached screenshot at that width.\n- A finding taken from the diff or the source instead of the running screen.\n- Taste the approved design does not state, reported as a finding.\n- Every deviation a Blocker — or none above Nitpick on a screen that visibly differs.\n","# Design Conformance\n\n## Overview\n\nA task built from an approved design is tested against that design too. A human approved it before any code was written, and its hand-off spec is the contract the developer built to — an oracle like the acceptance criteria and the analysis spec. You read the design documents, never the code (black-box-testing), and compare what the running product shows with what they say.\n\n**Core principle:** The approved design is the expected result. Same widths, same states, side by side — every finding cites the hand-off line or the mockup frame it contradicts, and the screenshot that shows it.\n\n## Inputs\n\n- **The hand-off spec** — `handoff: \u003Cscreen>`, first under \"## The approved design this task builds\": the components, the tokens, every state, the breakpoints, the exact copy. Its header names the chosen mockup, `design: \u003Cscreen> · \u003Cletter>`; the other variant documents were not chosen — never test against them.\n- **The chosen mockup** — its text rendition follows the hand-off in your context. To see a frame's look, render it under `$QA`, never the workspace: `list_task_documents \u003Cdesign key>` with `raw: true` (follow `next_offset` to the end), write it with a quoted heredoc (`mkdir -p \"$QA\u002Fdesign\"; cat > \"$QA\u002Fdesign\u002F\u003Cslug>.html\" \u003C\u003C'TT_HTML'` … `TT_HTML`), serve it on loopback (`nohup python3 -u -m http.server 0 --bind 127.0.0.1 --directory \"$QA\u002Fdesign\" > \"$QA\u002Fdesign.log\" 2>&1 & echo $! > \"$QA\u002Fdesign.pid\"`, the port is in the log), `browser_navigate` there and screenshot the frame you compare. Stop it by its PID (stop-what-you-started).\n- **The task's scope** — where the task description and the design disagree, the description is the narrower scope and wins for this task; a part of the design outside it is not a finding.\n\n## Cases\n\nAdd them to the matrix before you boot anything (scenario-plan-first): category `visual`, title prefix `design:` (`design: invoices empty 375 -> matches #a-375-empty`), linked to the criterion that names the state or the copy when there is one — the designer's hand-off criteria usually do. One case per state the hand-off lists, per width:\n\n- **Web:** 1440 (`browser_set_viewport {device:\"desktop\"}`), 768 (`{device:\"tablet\"}`), 375 (`{device:\"mobile\", width:375}`) — the design's widths. The four-width overflow check (360 · 768 · 1024 · 1440, ui-visual-evidence) still runs on its own.\n- **Mobile:** the phone frames on the device — portrait, and landscape where the hand-off draws it — and the tablet frames when the device or simulator is a tablet. A width you cannot reach is `skipped` with the reason, never assumed to match.\n- **States:** default, loading, empty, error, success, dark, the longest content — each one the hand-off lists, reached from the outside: an empty search, a filter with no result, an invalid submit, a failing dependency (frontend-manual-testing's `route` recipe). A state you cannot reach is `skipped`, \"not reachable from outside: \u003Cwhy>\".\n\n## Compare, in this order\n\n1. **Layout and hierarchy** — the regions, their order, what comes first, the one primary action.\n2. **Copy** — every string verbatim against the hand-off's Copy table; a paraphrase is a deviation.\n3. **Components** — the one the hand-off names, in its variant (`browser_read_dom` \u002F `mobile_read_ui` shows its role and label).\n4. **Tokens** — colour, type size, spacing, radius as the design system holds them (`get_design_system`); `browser_read_dom` with `as_text: false` shows the classes and styles an element carries. A colour the design system does not hold is a deviation.\n5. **States** — every listed state present, and drawn as designed.\n6. **Responsive behaviour** — per breakpoint the hand-off lists.\n7. **Visible accessibility it names** — labels, focus ring, status never colour alone, touch targets.\n\n## Evidence\n\nEvery comparison shot of the running UI goes on the task, where the human sees it: `browser_screenshot` (`full_page: true`, no `width` after `browser_set_viewport`) or `mobile_screenshot`, with `attach_to_task: true` and `title: \"\u003Cscreen>-\u003Cstate>-\u003Cwidth>\"` (`invoices-empty-375`). Cite it by that title in the case's `evidence` and the criterion note, with the URL, the viewport report and what it showed — a title, never a path.\n\n## Severity — what fails the round\n\n| Severity | Means | Example | Verdict |\n|---|---|---|---|\n| Blocker | the screen fails the approved design on a core path | the primary action missing at 375; the layout breaks at 768; body text fails contrast | case `failed`, criterion rejected, need_revision |\n| High | a deviation users will notice, or a design-system violation | the wrong component; an off-token colour; a missing empty state; key copy wrong | case `failed`, criterion rejected, need_revision |\n| Medium | noticeable, minor | spacing one step off; secondary copy paraphrased; alignment | case `passed`, the deviation in its `notes` and the criterion note |\n| Nitpick | polish | icon optical alignment; a 1px border difference | same as Medium |\n\nA Blocker or High fails the round like any other failure (bug-report-writing, qa-fail-to-need-revision): the case `failed` with `actual` = what the screen shows, the criterion it breaks rejected with expected (the hand-off line, or the frame id such as `#a-375-empty`) vs actual and the screenshot title, and one line in the numbered need_revision index — `3. [AC4 · case \"design: invoices empty 375\"] empty state has no \"New invoice\" action — High`. A failed design case with no criterion to reject still fails the round. Medium and Nitpick never fail the task: they are notes, like any polish beyond the floor.\n\nWrite each finding as a problem, not a prescription: what the design says, what the screen shows, where (URL + width + region) — never \"change the class\".\n\n## Red Flags\n\n- Testing against a variant document that was not chosen.\n- A `design:` case passed with no attached screenshot at that width.\n- A finding taken from the diff or the source instead of the running screen.\n- Taste the approved design does not state, reported as a finding.\n- Every deviation a Blocker — or none above Nitpick on a screen that visibly differs.\n",{"name":5,"category":24,"description":7},"qa","main","910394a0b6db9fc8c9a8f36328ec421d3ebd8b38","catalog\u002Fagents\u002Fqa-agent\u002Fskills\u002Fdesign-conformance",18,"https:\u002F\u002Fgithub.com\u002Fmakifbaysal\u002Ftasktrooper","https:\u002F\u002Ftasktrooper.ai","https:\u002F\u002Fgithub.com\u002Fmakifbaysal","Senior Software Developer","2026-11-15 04:50:29",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add makifbaysal\u002Ftasktrooper --skill design-conformance",{"tool":40,"command":41},"bunx","bunx skills add makifbaysal\u002Ftasktrooper --skill design-conformance",{"tool":43,"command":44},"pnpm","pnpm dlx skills add makifbaysal\u002Ftasktrooper --skill design-conformance","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmakifbaysal\u002Ftasktrooper\u002Fblob\u002Fmain\u002Fcatalog\u002Fagents\u002Fqa-agent\u002Fskills\u002Fdesign-conformance\u002FSKILL.md",true,[],[49,55,62,69,76,83],{"skillNo":50,"name":51,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M7CSVEPQW6J6TD6CS2T5FPNZ","visual-craft","https:\u002F\u002Fgithub.com\u002Fmakifbaysal\u002Ftasktrooper\u002Fblob\u002Fmain\u002Fcatalog\u002Fagents\u002Fui-designer\u002Fskills\u002Fvisual-craft\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"publishedAt":19},"01M7CSVAQG91ETHQ0HBNZDG5FR","ux-copy","Use when you write any words a user reads in a mockup or a hand-off - content language and voice, buttons, labels, errors, empty states, confirmations, numbers and dates, and the longest-string check","https:\u002F\u002Fgithub.com\u002Fmakifbaysal\u002Ftasktrooper\u002Fblob\u002Fmain\u002Fcatalog\u002Fagents\u002Fui-designer\u002Fskills\u002Fux-copy\u002FSKILL.md",[],[],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M7CSV6H0J38659NDXHJQ8396","screen-mockup-html","Use when you draw screens as the design task's HTML mockup - ONE canvas document holding every screen and every variant side by side, the sanitizer-safe recipe, phone and wide frames, the states grid, static control states, and the loopback self-check with screenshots at 375, 768 and 1440 attached to the task","https:\u002F\u002Fgithub.com\u002Fmakifbaysal\u002Ftasktrooper\u002Fblob\u002Fmain\u002Fcatalog\u002Fagents\u002Fui-designer\u002Fskills\u002Fscreen-mockup-html\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7CSV3JFRF3BM3XK52M7YEYR","performance-awareness","How the performance score moves for your role. Use when the score context shows a declining trend, or before handing off work you have not self-checked against every criterion.","https:\u002F\u002Fgithub.com\u002Fmakifbaysal\u002Ftasktrooper\u002Fblob\u002Fmain\u002Fcatalog\u002Fagents\u002Fui-designer\u002Fskills\u002Fperformance-awareness\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7CSTZDH00YRF5SCT7BRT9QD","design-variants","Use when a screen design needs variants - how many, variants as side-by-side rows on the one canvas, what makes two variants genuinely different, recommending one, how the human's choice is read, and narrowing to the chosen one in revision","https:\u002F\u002Fgithub.com\u002Fmakifbaysal\u002Ftasktrooper\u002Fblob\u002Fmain\u002Fcatalog\u002Fagents\u002Fui-designer\u002Fskills\u002Fdesign-variants\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7CSTV6P3SM3PAE28DB0RF77","design-system-authoring","Use when you derive, change or extend a design system - where tokens live per stack, the DTCG token tree, role-based names for both themes, base vs layer vs drift, the DESIGN.md section order, the inventory format, the contrast check and the lint every proposal returns","https:\u002F\u002Fgithub.com\u002Fmakifbaysal\u002Ftasktrooper\u002Fblob\u002Fmain\u002Fcatalog\u002Fagents\u002Fui-designer\u002Fskills\u002Fdesign-system-authoring\u002FSKILL.md",[],[]]