{
  "schemaVersion": "1.0",
  "item": {
    "slug": "youtube-thumbnail-design",
    "name": "Youtube Thumbnail Design",
    "source": "tencent",
    "type": "skill",
    "category": "内容创作",
    "sourceUrl": "https://clawhub.ai/okaris/youtube-thumbnail-design",
    "canonicalUrl": "https://clawhub.ai/okaris/youtube-thumbnail-design",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/youtube-thumbnail-design",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=youtube-thumbnail-design",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md"
    ],
    "primaryDoc": "SKILL.md",
    "quickSetup": [
      "Download the package from Yavira.",
      "Extract the archive and review SKILL.md first.",
      "Import or place the package into your OpenClaw setup."
    ],
    "agentAssist": {
      "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
      "steps": [
        "Download the package from Yavira.",
        "Extract it into a folder your agent can access.",
        "Paste one of the prompts below and point your agent at the extracted folder."
      ],
      "prompts": [
        {
          "label": "New install",
          "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
        },
        {
          "label": "Upgrade existing",
          "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
        }
      ]
    },
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-30T16:55:25.780Z",
      "expiresAt": "2026-05-07T16:55:25.780Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
        "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null
      },
      "scope": "source",
      "summary": "Source download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this source.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/youtube-thumbnail-design"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    },
    "downloadPageUrl": "https://openagent3.xyz/downloads/youtube-thumbnail-design",
    "agentPageUrl": "https://openagent3.xyz/skills/youtube-thumbnail-design/agent",
    "manifestUrl": "https://openagent3.xyz/skills/youtube-thumbnail-design/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/youtube-thumbnail-design/agent.md"
  },
  "agentAssist": {
    "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
    "steps": [
      "Download the package from Yavira.",
      "Extract it into a folder your agent can access.",
      "Paste one of the prompts below and point your agent at the extracted folder."
    ],
    "prompts": [
      {
        "label": "New install",
        "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
      },
      {
        "label": "Upgrade existing",
        "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "YouTube Thumbnail Design",
        "body": "Create high-CTR YouTube thumbnails with AI image generation via inference.sh CLI."
      },
      {
        "title": "Quick Start",
        "body": "curl -fsSL https://cli.inference.sh | sh && infsh login\n\n# Generate a thumbnail\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"YouTube thumbnail style, close-up of a person with surprised excited expression looking at a glowing laptop screen, vibrant blue and orange color scheme, dramatic studio lighting, shallow depth of field, high contrast, cinematic\",\n  \"width\": 1280,\n  \"height\": 720\n}'\n\nInstall note: The install script only detects your OS/architecture, downloads the matching binary from dist.inference.sh, and verifies its SHA-256 checksum. No elevated permissions or background processes. Manual install & verification available."
      },
      {
        "title": "Specifications",
        "body": "SpecValueDimensions1280 x 720 px (minimum)Recommended1920 x 1080 pxAspect ratio16:9Max file size2 MBFormatsJPG, GIF, PNG"
      },
      {
        "title": "The 120px Test",
        "body": "Your thumbnail appears at roughly 120px wide on mobile — that's how most viewers first see it.\n\nAt 120px, viewers must be able to identify:\n\nThe mood/emotion (from colors and expression)\nThe general subject (from composition)\nThe text (if any — only if large enough)\n\nTest: view your thumbnail at 120px width. If it's a muddy blur, redesign."
      },
      {
        "title": "Safe Zones",
        "body": "┌─────────────────────────────────────────────┐\n│                                             │\n│   ✅ SAFE FOR TEXT AND KEY ELEMENTS         │\n│                                             │\n│                                             │\n│                                             │\n│                                             │\n│                                       ┌───┐ │\n│                                       │ ⏱ │ │ ← Timestamp overlay\n│                              ┌────────┴───┘ │    (bottom-right)\n│   ┌────┐                     │  DURATION    │\n│   │ CH │ Chapter marker      └──────────────│\n└───┴────┴────────────────────────────────────┘\n     ↑ Bottom-left: chapter/progress markers\n\nAvoid placing critical elements in:\n\nBottom-right corner (video duration timestamp)\nBottom-left corner (chapter markers, progress bar)\nExtreme edges (cropping varies by device)"
      },
      {
        "title": "High-Contrast Pairs That Work",
        "body": "CombinationMoodBest ForYellow + BlackUrgency, attentionTech, business, listsRed + WhiteEnergy, excitementEntertainment, reactionsBlue + OrangeProfessional contrastEducation, tutorialsGreen + WhiteGrowth, moneyFinance, success storiesPurple + YellowPremium, creativeDesign, art, creativityWhite + DarkClean, minimalLuxury, minimalist channels"
      },
      {
        "title": "Color Rules",
        "body": "Background and text/subject should be complementary or high-contrast\nAvoid same-temperature colors touching (red on orange = mud)\nUse 3 colors maximum per thumbnail\nSaturate more than real life — thumbnails compete with bright UI"
      },
      {
        "title": "When to Use Text",
        "body": "Lists/numbers: \"7 Tips\", \"Top 10\"\nStrong opinions: \"STOP Doing This\"\nResults: \"$10K in 30 Days\"\nComparisons: \"vs\" between two things"
      },
      {
        "title": "When NOT to Use Text",
        "body": "The video title already says it (redundant)\nThe emotion/visual tells the story\nYou can't make it large enough to read at 120px"
      },
      {
        "title": "Text Rules",
        "body": "RuleReasonMax 6 wordsReadability at thumbnail sizeMin 60pt equivalentMust be legible at 120px widthBold sans-serif fontThin fonts disappear at small sizesContrast stroke/shadowEnsures readability on any backgroundNo small textIf it's not readable small, cut it"
      },
      {
        "title": "Face Expression Psychology",
        "body": "Thumbnails with faces get higher CTR than faceless thumbnails. Expression matters:\n\nExpressionCTR ImpactBest ForSurprise/shockHighestReaction, reveal, discovery contentCuriosityHighTutorial, how-to, tipsExcitementHighUnboxing, reviews, announcementsConcern/worryMedium-highWarning, mistake, problem contentConfidenceMediumExpert advice, authority contentNeutralLowestAvoid unless your brand is minimalist"
      },
      {
        "title": "Face Composition Rules",
        "body": "Face should fill 30-50% of the thumbnail\nEyes looking toward the text or subject (directs viewer attention)\nEyes looking at camera = connection. Eyes looking at object = curiosity.\nPlace face on one side (usually left), text or subject on the other\n\n# Generate a face-forward thumbnail\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"close-up portrait of a man with genuinely surprised expression, mouth slightly open, raised eyebrows, looking at camera, left side of frame, vibrant teal background, dramatic rim lighting, YouTube thumbnail style, high contrast, cinematic\",\n  \"width\": 1280,\n  \"height\": 720\n}'\n\n# Generate a face-looking-at-subject thumbnail\ninfsh app run bytedance/seedream-4-5 --input '{\n  \"prompt\": \"person looking amazed at a glowing holographic chart showing upward growth, dramatic blue and green lighting, right side profile view, dark background, tech aesthetic, high energy\",\n  \"size\": \"2K\"\n}'"
      },
      {
        "title": "Tutorial / How-To",
        "body": "infsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"overhead flat lay of organized workspace with laptop showing code editor, colorful sticky notes, coffee cup, clean bright background, professional setup, tutorial style composition, warm lighting\",\n  \"width\": 1280,\n  \"height\": 720\n}'"
      },
      {
        "title": "Before/After",
        "body": "infsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"split composition, left side dark and messy disorganized desk, right side bright clean organized minimalist workspace, dramatic contrast between chaos and order, clear dividing line in center, high contrast\",\n  \"width\": 1280,\n  \"height\": 720\n}'"
      },
      {
        "title": "Product Review / Comparison",
        "body": "infsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"two products facing each other with dramatic lighting and sparks between them, competition battle concept, dark background with colorful rim lighting, versus comparison style, high energy, product photography\",\n  \"width\": 1280,\n  \"height\": 720\n}'"
      },
      {
        "title": "Listicle / Number",
        "body": "infsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"dynamic arrangement of 7 different colorful objects floating in space against dark gradient background, each item distinct and clearly separated, energetic composition, vibrant saturated colors, studio lighting\",\n  \"width\": 1280,\n  \"height\": 720\n}'"
      },
      {
        "title": "A/B Testing",
        "body": "Test one variable at a time:\n\nVariableTest A vs BFace vs No faceSame composition, with/without personExpressionSurprise vs curiosityColor schemeWarm vs cool paletteText vs No textWith/without text overlayBackgroundBright vs darkCompositionLeft-facing vs right-facing subject\n\n# Generate variant A\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"..., bright yellow background, ...\",\n  \"width\": 1280, \"height\": 720\n}' --no-wait\n\n# Generate variant B (same prompt, different background)\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"..., dark navy background, ...\",\n  \"width\": 1280, \"height\": 720\n}' --no-wait"
      },
      {
        "title": "Thumbnail Checklist",
        "body": "1280x720 minimum (1920x1080 preferred)\n Under 2MB file size\n Passes the 120px squint test\n No critical elements in bottom-right (timestamp) or bottom-left (chapter)\n Max 3 colors, high contrast\n Text (if any) is max 6 words, bold, with contrast stroke\n Face expression matches content energy (if applicable)\n Doesn't duplicate the video title\n Stands out from surrounding thumbnails (check your niche)\n Works on both light and dark YouTube backgrounds"
      },
      {
        "title": "Common Mistakes",
        "body": "MistakeProblemFixToo much textUnreadable at thumbnail sizeMax 6 words or no textLow contrastDisappears in the feedUse complementary colorsCluttered compositionEye doesn't know where to lookOne focal pointGeneric stock photo feelNo personality, gets skippedAuthentic expressions, unique anglesTiny detailsLost at 120pxBold, simple shapesSame style every videoViewer fatigueVary within brand guidelinesMisleading thumbnailKills trust, hurts retentionMatch the actual content"
      },
      {
        "title": "Related Skills",
        "body": "npx skills add inference-sh/skills@ai-image-generation\nnpx skills add inference-sh/skills@image-upscaling\nnpx skills add inference-sh/skills@prompt-engineering\n\nBrowse all apps: infsh app list"
      }
    ],
    "body": "YouTube Thumbnail Design\n\nCreate high-CTR YouTube thumbnails with AI image generation via inference.sh CLI.\n\nQuick Start\ncurl -fsSL https://cli.inference.sh | sh && infsh login\n\n# Generate a thumbnail\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"YouTube thumbnail style, close-up of a person with surprised excited expression looking at a glowing laptop screen, vibrant blue and orange color scheme, dramatic studio lighting, shallow depth of field, high contrast, cinematic\",\n  \"width\": 1280,\n  \"height\": 720\n}'\n\n\nInstall note: The install script only detects your OS/architecture, downloads the matching binary from dist.inference.sh, and verifies its SHA-256 checksum. No elevated permissions or background processes. Manual install & verification available.\n\nSpecifications\nSpec\tValue\nDimensions\t1280 x 720 px (minimum)\nRecommended\t1920 x 1080 px\nAspect ratio\t16:9\nMax file size\t2 MB\nFormats\tJPG, GIF, PNG\nThe 120px Test\n\nYour thumbnail appears at roughly 120px wide on mobile — that's how most viewers first see it.\n\nAt 120px, viewers must be able to identify:\n\nThe mood/emotion (from colors and expression)\nThe general subject (from composition)\nThe text (if any — only if large enough)\n\nTest: view your thumbnail at 120px width. If it's a muddy blur, redesign.\n\nSafe Zones\n┌─────────────────────────────────────────────┐\n│                                             │\n│   ✅ SAFE FOR TEXT AND KEY ELEMENTS         │\n│                                             │\n│                                             │\n│                                             │\n│                                             │\n│                                       ┌───┐ │\n│                                       │ ⏱ │ │ ← Timestamp overlay\n│                              ┌────────┴───┘ │    (bottom-right)\n│   ┌────┐                     │  DURATION    │\n│   │ CH │ Chapter marker      └──────────────│\n└───┴────┴────────────────────────────────────┘\n     ↑ Bottom-left: chapter/progress markers\n\n\nAvoid placing critical elements in:\n\nBottom-right corner (video duration timestamp)\nBottom-left corner (chapter markers, progress bar)\nExtreme edges (cropping varies by device)\nColor Strategy\nHigh-Contrast Pairs That Work\nCombination\tMood\tBest For\nYellow + Black\tUrgency, attention\tTech, business, lists\nRed + White\tEnergy, excitement\tEntertainment, reactions\nBlue + Orange\tProfessional contrast\tEducation, tutorials\nGreen + White\tGrowth, money\tFinance, success stories\nPurple + Yellow\tPremium, creative\tDesign, art, creativity\nWhite + Dark\tClean, minimal\tLuxury, minimalist channels\nColor Rules\nBackground and text/subject should be complementary or high-contrast\nAvoid same-temperature colors touching (red on orange = mud)\nUse 3 colors maximum per thumbnail\nSaturate more than real life — thumbnails compete with bright UI\nText on Thumbnails\nWhen to Use Text\nLists/numbers: \"7 Tips\", \"Top 10\"\nStrong opinions: \"STOP Doing This\"\nResults: \"$10K in 30 Days\"\nComparisons: \"vs\" between two things\nWhen NOT to Use Text\nThe video title already says it (redundant)\nThe emotion/visual tells the story\nYou can't make it large enough to read at 120px\nText Rules\nRule\tReason\nMax 6 words\tReadability at thumbnail size\nMin 60pt equivalent\tMust be legible at 120px width\nBold sans-serif font\tThin fonts disappear at small sizes\nContrast stroke/shadow\tEnsures readability on any background\nNo small text\tIf it's not readable small, cut it\nFace Expression Psychology\n\nThumbnails with faces get higher CTR than faceless thumbnails. Expression matters:\n\nExpression\tCTR Impact\tBest For\nSurprise/shock\tHighest\tReaction, reveal, discovery content\nCuriosity\tHigh\tTutorial, how-to, tips\nExcitement\tHigh\tUnboxing, reviews, announcements\nConcern/worry\tMedium-high\tWarning, mistake, problem content\nConfidence\tMedium\tExpert advice, authority content\nNeutral\tLowest\tAvoid unless your brand is minimalist\nFace Composition Rules\nFace should fill 30-50% of the thumbnail\nEyes looking toward the text or subject (directs viewer attention)\nEyes looking at camera = connection. Eyes looking at object = curiosity.\nPlace face on one side (usually left), text or subject on the other\n# Generate a face-forward thumbnail\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"close-up portrait of a man with genuinely surprised expression, mouth slightly open, raised eyebrows, looking at camera, left side of frame, vibrant teal background, dramatic rim lighting, YouTube thumbnail style, high contrast, cinematic\",\n  \"width\": 1280,\n  \"height\": 720\n}'\n\n# Generate a face-looking-at-subject thumbnail\ninfsh app run bytedance/seedream-4-5 --input '{\n  \"prompt\": \"person looking amazed at a glowing holographic chart showing upward growth, dramatic blue and green lighting, right side profile view, dark background, tech aesthetic, high energy\",\n  \"size\": \"2K\"\n}'\n\nThumbnail Patterns by Content Type\nTutorial / How-To\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"overhead flat lay of organized workspace with laptop showing code editor, colorful sticky notes, coffee cup, clean bright background, professional setup, tutorial style composition, warm lighting\",\n  \"width\": 1280,\n  \"height\": 720\n}'\n\nBefore/After\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"split composition, left side dark and messy disorganized desk, right side bright clean organized minimalist workspace, dramatic contrast between chaos and order, clear dividing line in center, high contrast\",\n  \"width\": 1280,\n  \"height\": 720\n}'\n\nProduct Review / Comparison\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"two products facing each other with dramatic lighting and sparks between them, competition battle concept, dark background with colorful rim lighting, versus comparison style, high energy, product photography\",\n  \"width\": 1280,\n  \"height\": 720\n}'\n\nListicle / Number\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"dynamic arrangement of 7 different colorful objects floating in space against dark gradient background, each item distinct and clearly separated, energetic composition, vibrant saturated colors, studio lighting\",\n  \"width\": 1280,\n  \"height\": 720\n}'\n\nA/B Testing\n\nTest one variable at a time:\n\nVariable\tTest A vs B\nFace vs No face\tSame composition, with/without person\nExpression\tSurprise vs curiosity\nColor scheme\tWarm vs cool palette\nText vs No text\tWith/without text overlay\nBackground\tBright vs dark\nComposition\tLeft-facing vs right-facing subject\n# Generate variant A\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"..., bright yellow background, ...\",\n  \"width\": 1280, \"height\": 720\n}' --no-wait\n\n# Generate variant B (same prompt, different background)\ninfsh app run falai/flux-dev-lora --input '{\n  \"prompt\": \"..., dark navy background, ...\",\n  \"width\": 1280, \"height\": 720\n}' --no-wait\n\nThumbnail Checklist\n 1280x720 minimum (1920x1080 preferred)\n Under 2MB file size\n Passes the 120px squint test\n No critical elements in bottom-right (timestamp) or bottom-left (chapter)\n Max 3 colors, high contrast\n Text (if any) is max 6 words, bold, with contrast stroke\n Face expression matches content energy (if applicable)\n Doesn't duplicate the video title\n Stands out from surrounding thumbnails (check your niche)\n Works on both light and dark YouTube backgrounds\nCommon Mistakes\nMistake\tProblem\tFix\nToo much text\tUnreadable at thumbnail size\tMax 6 words or no text\nLow contrast\tDisappears in the feed\tUse complementary colors\nCluttered composition\tEye doesn't know where to look\tOne focal point\nGeneric stock photo feel\tNo personality, gets skipped\tAuthentic expressions, unique angles\nTiny details\tLost at 120px\tBold, simple shapes\nSame style every video\tViewer fatigue\tVary within brand guidelines\nMisleading thumbnail\tKills trust, hurts retention\tMatch the actual content\nRelated Skills\nnpx skills add inference-sh/skills@ai-image-generation\nnpx skills add inference-sh/skills@image-upscaling\nnpx skills add inference-sh/skills@prompt-engineering\n\n\nBrowse all apps: infsh app list"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/okaris/youtube-thumbnail-design",
    "publisherUrl": "https://clawhub.ai/okaris/youtube-thumbnail-design",
    "owner": "okaris",
    "version": "0.1.5",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/youtube-thumbnail-design",
    "downloadUrl": "https://openagent3.xyz/downloads/youtube-thumbnail-design",
    "agentUrl": "https://openagent3.xyz/skills/youtube-thumbnail-design/agent",
    "manifestUrl": "https://openagent3.xyz/skills/youtube-thumbnail-design/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/youtube-thumbnail-design/agent.md"
  }
}