[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MJMRDWZ4KTDFTHPD3R72CM":3,"skill-same:01M6MJMRDWZ4KTDFTHPD3R72CM":80},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":30,"publishedAt":51,"linkOnly":52,"markdownContent":53,"markdownBody":54,"frontMatter":55,"refName":60,"commitSha":61,"skillDirectory":62,"forksCount":13,"repositoryHtmlUrl":63,"repositoryHomepageUrl":6,"ownerHtmlUrl":64,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":65,"install":66,"resources":79},"01M6MJMRDWZ4KTDFTHPD3R72CM","apple-hig-expert",null,"Audits and designs iOS\u002FmacOS\u002FwatchOS\u002FvisionOS interfaces against the Apple Human Interface Guidelines, including the Liquid Glass design language (announced WWDC25, shipped with iOS 26\u002FmacOS Tahoe, Sept 2025). Use when reviewing an Apple-platform mockup or app for HIG compliance, checking contrast or tap-target sizes, or designing native-feeling Apple UI (e.g., 'audit my iOS app against the HIG', 'is this text readable on Liquid Glass?').","JiayiV\u002Fquestion_mixing","JiayiV","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F54466889?v=4","01M6JFJYYMRZ3QRGP9339JNQ2K","01M6JFJZ35KB0QVT50NAHPJA7G",0,"MIT","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fpublic-02-claude-skills-main-product-team-apple-hig-expert-skills-apple-hig-expert-skill-54ce0884\u002FSKILL.md",[17,22,27],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000015","视觉设计","SCENARIO","使用场景",{"categoryNo":23,"categoryName":24,"taxonomyType":25,"taxonomyTypeDesc":26},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",{"categoryNo":28,"categoryName":29,"taxonomyType":25,"taxonomyTypeDesc":26},"01JBSEEDCAT000000000000037","律师与法务",[31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50],"apple hig","apple hig expert","expert","apple","for","design","design language","interface","with","2025","native","app","platform","ios app","tap","liquid glass design","use","text","liquid glass","guidelines","2026-11-10 10:12:30",false,"---\nname: apple-hig-expert\ndescription: \"Audits and designs iOS\u002FmacOS\u002FwatchOS\u002FvisionOS interfaces against the Apple Human Interface Guidelines, including the Liquid Glass design language (announced WWDC25, shipped with iOS 26\u002FmacOS Tahoe, Sept 2025). Use when reviewing an Apple-platform mockup or app for HIG compliance, checking contrast or tap-target sizes, or designing native-feeling Apple UI (e.g., 'audit my iOS app against the HIG', 'is this text readable on Liquid Glass?').\"\nlicense: MIT\nmetadata:\n  version: 1.1.0\n  author: Alireza Rezvani\n  category: design\n  updated: 2026-06-11\n---\n\n# Apple HIG Expert\n\nDesign and audit apps against the Apple Human Interface Guidelines (HIG, [developer.apple.com\u002Fdesign\u002Fhuman-interface-guidelines](https:\u002F\u002Fdeveloper.apple.com\u002Fdesign\u002Fhuman-interface-guidelines)), including the **Liquid Glass** design language. HIG content evolves with each OS release — when a claim matters, verify against the live HIG pages cited in `references\u002F`.\n\n## Before Starting\n\nIf `product-context.md` or `ios-design-context.md` exists, read it before asking questions. Then gather:\n\n1. **Platform target**: iOS, macOS, watchOS, or visionOS?\n2. **Current state**: new design or auditing an existing mockup\u002Fcode?\n3. **App category**: utility, productivity, game, social, etc.\n\n## Modes\n\n- **Mode 1 — Design from scratch**: pick the platform navigation paradigm and layout primitives first (see `references\u002Fplatform-specifics.md`), then apply typography and semantic color (`references\u002Fvisual-design.md`).\n- **Mode 2 — HIG audit**: fill in `templates\u002Fhig-audit-template.md`, run `scripts\u002Fhig_checker.py` on every measurable element, and deliver a scored report (see Worked example below).\n\n## The Compliance Tool\n\n`scripts\u002Fhig_checker.py` (stdlib-only) has three subcommands:\n\n```bash\n# 1. Contrast ratio (WCAG formula; pass >= 4.5:1 for normal text)\npython3 scripts\u002Fhig_checker.py contrast \"#8E8E93\" \"#FFFFFF\"\n# -> Contrast Ratio: 3.26 [FAILED]\n\n# 2. Tap-target size (pass >= 44x44 pt per HIG)\npython3 scripts\u002Fhig_checker.py target 32 32\n# -> Tap Target: 32x32 [FAILED]\n\n# 3. Batch audit from JSON -> scorecard (starts at 100, -10 per violation)\npython3 scripts\u002Fhig_checker.py batch audit.json\n```\n\nBatch input shape:\n\n```json\n{\n  \"checks\": [\n    {\"type\": \"contrast\", \"name\": \"caption-on-card\", \"fg\": \"#8E8E93\", \"bg\": \"#FFFFFF\"},\n    {\"type\": \"target\", \"name\": \"close-button\", \"w\": 32, \"h\": 32}\n  ]\n}\n```\n\n**Scorecard rubric:** the batch score starts at 100 and subtracts 10 per failed check; violations are listed by element name. 90-100 = ship, 70-80 = fix before release, below 70 = systematic rework. Checks the tool cannot measure (VoiceOver labels, Dynamic Type behavior, Reduce Transparency) are assessed manually via the audit template and tagged with confidence.\n\n## Worked example: iOS settings-screen audit\n\n**Input:** mockup with body text `#1C1C1E` and captions `#8E8E93` on white cards, a 32x32 pt close button, and a 343x50 pt primary CTA.\n\n**Run:**\n\n```bash\npython3 scripts\u002Fhig_checker.py batch audit.json\n```\n\n**Output (real):**\n\n```json\n{\n  \"score\": 80,\n  \"violations\": [\n    \"Contrast 3.26 fails for caption-on-card\",\n    \"Target 32x32 small for close-button\"\n  ]\n}\n```\n\n**Findings → fixes (bottom line first):**\n\n> **HIG score 80\u002F100 — two fixes before release.**\n> 1. Captions fail contrast (3.26 \u003C 4.5). Use `.secondaryLabel` (semantic color) instead of hardcoded `#8E8E93`, or darken to ≥ `#6E6E73` on white. ? verified by tool.\n> 2. Close button is 32x32 pt (\u003C 44x44 minimum). Keep the glyph small but expand the hit region to 44x44 with padding\u002F`contentShape`. ? verified by tool.\n> 3. Manual check: the card uses an ultra-thin material over a photo background — re-test caption contrast against the *busiest* underlying region and with Reduce Transparency on. ? needs device test.\n\n## Core Design Principles\n\n1. **Liquid Glass** — translucent material hierarchy (announced at WWDC25, June 2025; shipped Sept 2025 across iOS 26, iPadOS 26, macOS Tahoe, watchOS 26, tvOS 26, visionOS 26). In SwiftUI, apply it via the `glassEffect` view modifier; keep hierarchy between content and controls. See `references\u002Fvisual-design.md`.\n2. **Accessibility first** — VoiceOver labels on every element, 44x44 pt minimum targets, 4.5:1 contrast for normal text (3:1 large text), Dynamic Type support. See `references\u002Faccessibility.md`.\n3. **Platform ergonomics** — tab bars\u002Fthumb reach on iOS, sidebars + menu bar + shortcuts on macOS, ornaments + gaze states on visionOS, glanceable vertical layouts on watchOS. See `references\u002Fplatform-specifics.md`.\n\n## Proactive Triggers\n\nSurface these WITHOUT being asked: low contrast over translucent layers; interactive elements under 44 pt; icon buttons with no accessibility label; density overload (no breathing room between glass layers).\n\n## Communication\n\n- **Bottom line first** — compliance status before details.\n- **What + Why + How** — \"Expand the hit region (What) because 32 pt targets fail the HIG minimum (Why); pad to 44x44 via contentShape (How).\"\n- **Confidence tagging** — ? tool-verified \u002F ? needs device test \u002F ? assumed.\n\n## Related Skills\n\n- **ui-design-system**: token-based component systems (not platform HIG rules).\n- **ux-researcher-designer**: persona\u002Fresearch validation (not visual styling).\n- **landing-page-generator**: web marketing pages, not native apps.\n","# Apple HIG Expert\n\nDesign and audit apps against the Apple Human Interface Guidelines (HIG, [developer.apple.com\u002Fdesign\u002Fhuman-interface-guidelines](https:\u002F\u002Fdeveloper.apple.com\u002Fdesign\u002Fhuman-interface-guidelines)), including the **Liquid Glass** design language. HIG content evolves with each OS release — when a claim matters, verify against the live HIG pages cited in `references\u002F`.\n\n## Before Starting\n\nIf `product-context.md` or `ios-design-context.md` exists, read it before asking questions. Then gather:\n\n1. **Platform target**: iOS, macOS, watchOS, or visionOS?\n2. **Current state**: new design or auditing an existing mockup\u002Fcode?\n3. **App category**: utility, productivity, game, social, etc.\n\n## Modes\n\n- **Mode 1 — Design from scratch**: pick the platform navigation paradigm and layout primitives first (see `references\u002Fplatform-specifics.md`), then apply typography and semantic color (`references\u002Fvisual-design.md`).\n- **Mode 2 — HIG audit**: fill in `templates\u002Fhig-audit-template.md`, run `scripts\u002Fhig_checker.py` on every measurable element, and deliver a scored report (see Worked example below).\n\n## The Compliance Tool\n\n`scripts\u002Fhig_checker.py` (stdlib-only) has three subcommands:\n\n```bash\n# 1. Contrast ratio (WCAG formula; pass >= 4.5:1 for normal text)\npython3 scripts\u002Fhig_checker.py contrast \"#8E8E93\" \"#FFFFFF\"\n# -> Contrast Ratio: 3.26 [FAILED]\n\n# 2. Tap-target size (pass >= 44x44 pt per HIG)\npython3 scripts\u002Fhig_checker.py target 32 32\n# -> Tap Target: 32x32 [FAILED]\n\n# 3. Batch audit from JSON -> scorecard (starts at 100, -10 per violation)\npython3 scripts\u002Fhig_checker.py batch audit.json\n```\n\nBatch input shape:\n\n```json\n{\n  \"checks\": [\n    {\"type\": \"contrast\", \"name\": \"caption-on-card\", \"fg\": \"#8E8E93\", \"bg\": \"#FFFFFF\"},\n    {\"type\": \"target\", \"name\": \"close-button\", \"w\": 32, \"h\": 32}\n  ]\n}\n```\n\n**Scorecard rubric:** the batch score starts at 100 and subtracts 10 per failed check; violations are listed by element name. 90-100 = ship, 70-80 = fix before release, below 70 = systematic rework. Checks the tool cannot measure (VoiceOver labels, Dynamic Type behavior, Reduce Transparency) are assessed manually via the audit template and tagged with confidence.\n\n## Worked example: iOS settings-screen audit\n\n**Input:** mockup with body text `#1C1C1E` and captions `#8E8E93` on white cards, a 32x32 pt close button, and a 343x50 pt primary CTA.\n\n**Run:**\n\n```bash\npython3 scripts\u002Fhig_checker.py batch audit.json\n```\n\n**Output (real):**\n\n```json\n{\n  \"score\": 80,\n  \"violations\": [\n    \"Contrast 3.26 fails for caption-on-card\",\n    \"Target 32x32 small for close-button\"\n  ]\n}\n```\n\n**Findings → fixes (bottom line first):**\n\n> **HIG score 80\u002F100 — two fixes before release.**\n> 1. Captions fail contrast (3.26 \u003C 4.5). Use `.secondaryLabel` (semantic color) instead of hardcoded `#8E8E93`, or darken to ≥ `#6E6E73` on white. ? verified by tool.\n> 2. Close button is 32x32 pt (\u003C 44x44 minimum). Keep the glyph small but expand the hit region to 44x44 with padding\u002F`contentShape`. ? verified by tool.\n> 3. Manual check: the card uses an ultra-thin material over a photo background — re-test caption contrast against the *busiest* underlying region and with Reduce Transparency on. ? needs device test.\n\n## Core Design Principles\n\n1. **Liquid Glass** — translucent material hierarchy (announced at WWDC25, June 2025; shipped Sept 2025 across iOS 26, iPadOS 26, macOS Tahoe, watchOS 26, tvOS 26, visionOS 26). In SwiftUI, apply it via the `glassEffect` view modifier; keep hierarchy between content and controls. See `references\u002Fvisual-design.md`.\n2. **Accessibility first** — VoiceOver labels on every element, 44x44 pt minimum targets, 4.5:1 contrast for normal text (3:1 large text), Dynamic Type support. See `references\u002Faccessibility.md`.\n3. **Platform ergonomics** — tab bars\u002Fthumb reach on iOS, sidebars + menu bar + shortcuts on macOS, ornaments + gaze states on visionOS, glanceable vertical layouts on watchOS. See `references\u002Fplatform-specifics.md`.\n\n## Proactive Triggers\n\nSurface these WITHOUT being asked: low contrast over translucent layers; interactive elements under 44 pt; icon buttons with no accessibility label; density overload (no breathing room between glass layers).\n\n## Communication\n\n- **Bottom line first** — compliance status before details.\n- **What + Why + How** — \"Expand the hit region (What) because 32 pt targets fail the HIG minimum (Why); pad to 44x44 via contentShape (How).\"\n- **Confidence tagging** — ? tool-verified \u002F ? needs device test \u002F ? assumed.\n\n## Related Skills\n\n- **ui-design-system**: token-based component systems (not platform HIG rules).\n- **ux-researcher-designer**: persona\u002Fresearch validation (not visual styling).\n- **landing-page-generator**: web marketing pages, not native apps.\n",{"name":5,"license":14,"metadata":56,"description":7},{"author":57,"updated":58,"version":59,"category":36},"Alireza Rezvani",1781136000000,"1.1.0","main","068c2486ccc0f8d200a3db4d21ec137064eda3b4","skills\u002Fpublic-02-claude-skills-main-product-team-apple-hig-expert-skills-apple-hig-expert-skill-54ce0884","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing","https:\u002F\u002Fgithub.com\u002FJiayiV","2026-08-25 16:30:52",{"source":8,"skillName":5,"commands":67,"prompt":77,"zipAvailable":78},[68,71,74],{"tool":69,"command":70},"npx","npx skills add JiayiV\u002Fquestion_mixing --skill apple-hig-expert",{"tool":72,"command":73},"bunx","bunx skills add JiayiV\u002Fquestion_mixing --skill apple-hig-expert",{"tool":75,"command":76},"pnpm","pnpm dlx skills add JiayiV\u002Fquestion_mixing --skill apple-hig-expert","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fpublic-02-claude-skills-main-product-team-apple-hig-expert-skills-apple-hig-expert-skill-54ce0884\u002FSKILL.md",true,[],[81,104,132,163,191,224],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":51},"01M6NN9PJS9NHWPDG4QP93BP8F","webmcpify","Make a web app agent-ready — propose a WebMCP tool manifest, integrate, verify in a real browser, heal; unrelated code stays untouched. Use for \"webmcpify\", \"add WebMCP\", or \"expose app actions to AI agents\".","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fwebmcpify\u002FSKILL.md",[],[83,88,89,90,91,92,93,94,95,96,97,98,99,47,100,101,42,102,35,103],"github","ai agents","web","notes","code","browser","https","actions","manifest","inventory","verify","tool","ready","unrelated","scope","webmcp",{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":108,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":111,"publishedAt":51},"01M6NN6KZE5RBN6NGHN2WXV1RE","vardoger-analyze","Use when the user asks to personalize the GitHub Copilot CLI assistant, adapt Copilot to their style, use vardoger, or analyze their Copilot CLI conversation history. Reads the local session directory at `~\u002F.copilot\u002Fsession-state\u002F`, extracts recurring preferences and conventions, and writes a fenced personalization block into `~\u002F.copilot\u002Fcopilot-instructions.md`. Runs entirely on the user's machine via the local `vardoger` CLI (`pipx install vardoger`); no network calls and no uploads. Triggers: 'personalize my copilot', 'analyze my copilot history', 'tailor copilot to me', 'run vardoger', 'update my copilot instructions from history', 'make copilot learn my style'.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fvardoger-analyze\u002FSKILL.md",[],[112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131],"vardoger analyze","analyze","directory","cli","github copilot","copilot","run","update","learn","network","my style","install","conversation","me","local","github copilot cli","pipx","my","at","copilot instructions",{"skillNo":133,"name":134,"nameZh":6,"description":135,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":136,"categories":137,"keywords":144,"publishedAt":51},"01M6NN3CZGQP3XAJ04BM1NNYGT","transloadit-media-processing","Process media files (video, audio, images, documents) using Transloadit. Use when asked to encode video to HLS\u002FMP4, generate thumbnails, resize or watermark images, extract audio, concatenate clips, add subtitles, OCR documents, or run any media processing pipeline. Covers 86+ processing robots for file transformation at scale.","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Ftransloadit-media-processing\u002FSKILL.md",[138,141],{"categoryNo":139,"categoryName":140,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":142,"categoryName":143,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发",[145,146,147,148,149,150,151,152,153,154,155,156,157,47,158,159,160,161,35,162],"transloadit media processing","Media","media processing","processing","MCP","mcp server","documents","subtitles","pipeline","ocr","Video","Audio","mp4","images","files","transformation","free","hls",{"skillNo":164,"name":165,"nameZh":6,"description":166,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":167,"categories":168,"keywords":172,"publishedAt":51},"01M6NN28E92F9GH4WFYQ2H1Y15","terraform-azurerm-set-diff-analyzer","Analyze Terraform plan JSON output for AzureRM Provider to distinguish between false-positive diffs (order-only changes in Set-type attributes) and actual resource changes. Use when reviewing terraform plan output for Azure resources like Application Gateway, Load Balancer, Firewall, Front Door, NSG, and other resources with Set-type attributes that cause spurious diffs due to internal ordering changes.","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fterraform-azurerm-set-diff-analyzer\u002FSKILL.md",[169],{"categoryNo":170,"categoryName":171,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000011","运维与部署",[173,174,175,176,177,178,179,180,181,47,182,183,184,185,186,187,188,189,190,35],"terraform","SET","terraform azurerm set diff analyzer","diff","analyzer","diffs","azure","gateway","firewall","only","resource changes","provider","resources","load balancer","attributes","front","load","balancer",{"skillNo":192,"name":193,"nameZh":6,"description":194,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":195,"categories":196,"keywords":206,"publishedAt":51},"01M6NN1QMH2TDYWTH3XY3N5XPV","technical-job-search","Use this skill when a software engineer asks for help with job search tasks: parsing or analyzing a job description, tailoring a CV\u002Fresume, writing a cover letter, evaluating a job offer, or drafting a post-interview follow-up email. Do not activate for general career advice unrelated to an active job search action.","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Ftechnical-job-search\u002FSKILL.md",[197,200,203],{"categoryNo":198,"categoryName":199,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":201,"categoryName":202,"taxonomyType":25,"taxonomyTypeDesc":26},"01JBSEEDCAT000000000000036","内容创作者",{"categoryNo":204,"categoryName":205,"taxonomyType":25,"taxonomyTypeDesc":26},"01JBSEEDCAT000000000000038","人力资源",[207,208,209,210,211,212,213,214,113,215,47,216,217,218,219,220,221,222,35,223],"search","Job Search","technical job search","technical","writing","tasks","email","parsing","interview","up","engineer","general","activate","do","task","cv","letter",{"skillNo":225,"name":226,"nameZh":6,"description":227,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":228,"categories":229,"keywords":230,"publishedAt":51},"01M6NMZFA67XYS1GS09WT42235","steno-mode","Shorthand-first response compression that cuts ~40% of response tokens while preserving technical precision and exact literals. Use when the user says \"steno mode\", \"shorthand mode\", \"compressed responses\", \"token reduction\", \"brief structured output\", or invokes \u002Fsteno. Supports four compression levels: lite, brief, court, machine. Do not trigger for requests needing polished prose such as onboarding\u002Ftutorial content, stakeholder or customer-facing copy, or teaching-focused explanations.","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fsteno-mode\u002FSKILL.md",[],[231,232,233,234,235,210,236,47,220,237,238,239,240,241,242,243,244,245,35,246],"steno mode","mode","customer","brief","first","tutorial","teaching","tokens","as","trigger","structured output","machine","output","court","requests","response compression"]