Exact request sent to the run panel.
{
"source": "retrieve",
"template": {
"id": "tiktok_personality",
"slug": "tiktok-brain-scan",
"name": "TikTok Brain Scan",
"categoryId": "ROAST_VIBE"
},
"inputs": {
"target": "",
"videoCount": 15,
"normalizedTarget": {
"raw": ""
}
},
"agentHint": "You are analyzing a TikTok profile with visual thumbnail analysis. Adapt to the UI.\n\n**TASK CONFIGURATION:**\n- Target: \n- Videos to Analyze: 15\n\n---\n\n## STEP 1: NAVIGATE TO PROFILE\n\n1. Parse target - extract username\n2. Navigate to tiktok.com/@[username]\n3. Wait for profile to fully load\n4. **Check accessibility:**\n - Profile exists? If not, report error\n - Private account? Report limitation\n - Videos visible? Proceed if yes\n\n---\n\n## STEP 2: EXTRACT PROFILE METADATA\n\n### Bio & Identity\n- Display name\n- Username\n- Bio text\n- Links (Linktree, etc.)\n- Verified status\n- Category/creator type label\n\n### Metrics\n- Follower count\n- Following count\n- Total likes\n- Video count\n\n### Profile Picture\n- Describe the profile picture\n- Style and vibe\n\n---\n\n## STEP 3: EXTRACT VIDEOS WITH THUMBNAILS\n\n**Extract 15 videos. For each capture:**\n\n1. **Thumbnail image URL** - the video preview/cover image\n2. **Video title/caption** - if visible\n3. **View count** - engagement indicator\n4. **Likes/comments** - if visible on hover\n5. **Duration** - video length\n6. **Sound/audio name** - if visible\n7. **Pinned status** - is this pinned?\n\n**Store structured:**\n```\nvideos: [\n {\n thumbnailUrl: \"url\",\n caption: \"text\",\n views: X,\n likes: Y,\n duration: \"0:15\",\n sound: \"sound name if visible\",\n isPinned: false\n }\n]\n```\n\n---\n\n## STEP 4: ANALYZE EACH THUMBNAIL (CRITICAL)\n\n**For each video thumbnail, generate detailed visual analysis:**\n\n### Describe:\n- **Scene/setting**: Where is this filmed? (bedroom, outdoor, studio, car, etc.)\n- **Person presence**: Face visible? Full body? Hands only? No person?\n- **Expression/pose**: If person visible - expression, energy, pose\n- **Text overlays**: Any text on thumbnail, what does it say?\n- **Visual hooks**: What makes this clickable? Curiosity gap?\n- **Lighting/quality**: Ring light, natural, low quality, professional\n- **Editing indicators**: Heavy effects visible in thumbnail?\n- **Format type**: Talking head, skit, dance, tutorial, reaction, duet, stitch\n- **Aesthetic**: Chaotic, polished, raw, curated\n- **Trend indicators**: Recognizable trend formats or templates\n\n### Pattern recognition across thumbnails:\n- Consistent visual style or random?\n- Face-forward thumbnails vs. not\n- Text-heavy vs. visual-only\n- Signature elements that repeat\n\n---\n\n## STEP 5: ANALYZE CONTENT PATTERNS\n\n### Video Format Mix\nBased on thumbnails, categorize:\n- Talking head %\n- Skits/comedy %\n- Dance/performance %\n- Tutorial/educational %\n- Reactions/duets %\n- Lifestyle/vlog %\n- Trend participation %\n\n### Visual Consistency\n- Same filming location recurring?\n- Consistent lighting setup?\n- Similar editing style?\n- Recognizable \"brand\" elements?\n\n### Thumbnail Strategy\n- Are thumbnails optimized for clicks?\n- Text overlay patterns\n- Expression patterns\n- Hook strategies visible\n\n---\n\n## STEP 6: AUDIO/TREND ANALYSIS\n\n**From visible sound names:**\n- Original sounds vs. trending sounds\n- Music genres favored\n- Trend participation rate\n- Sound originality score\n\n---\n\n## STEP 7: ENGAGEMENT PATTERN ANALYSIS\n\n**From view/like counts:**\n- Average performance\n- Viral hits (outliers)\n- Consistent performers\n- What visually differs in top performers?\n\n---\n\n## STEP 8: GENERATE BRAIN SCAN\n\n### 1. CREATOR ARCHETYPE\nBased on visual analysis, classify:\n- Examples: \"Reply Guy Energy\", \"Main Character\", \"Unhinged But Lovable\", \"Corporate TikTok Escapee\", \"Dance Trend Devotee\", \"Chaos Goblin\", \"Tutorial Queen\", \"Thirst Trap Artist\", \"Storytimer\", \"POV Actor\"\n- Must be based on ACTUAL visual patterns\n\n### 2. CONTENT FORMULA\n- What's their visual schtick?\n- Recurring thumbnail elements\n- Signature style markers\n\n### 3. TREND-CHASING SCORE (0-100)\nBased on visual analysis:\n- Original content indicators vs. trend templates\n- Format originality\n- Sound choices\n\n### 4. VISUAL STYLE ASSESSMENT\n- Filming quality and setup\n- Editing intensity\n- Thumbnail optimization level\n- Production value spectrum\n\n### 5. THUMBNAIL GAME RATING\n- How clickable are their covers?\n- Text overlay effectiveness\n- Expression optimization\n- Curiosity gap usage\n\n### 6. THE ALGORITHM KNOWS\nBased on visual patterns, predict:\n- What TikTok's algorithm thinks about them\n- Content categorization\n- Audience demographic guess\n\n### 7. VIRAL POTENTIAL ASSESSMENT\n- What visual elements are working\n- What's holding them back\n- Specific thumbnail/visual improvements\n\n---\n\n## STEP 9: REPORT RESULTS\n\n```json\n{\n \"target\": \"\",\n \"username\": \"\",\n \"profileMetrics\": {\n \"followers\": <number>,\n \"following\": <number>,\n \"totalLikes\": <number>,\n \"videoCount\": <number>\n },\n \"videosAnalyzed\": {\n \"count\": <number>,\n \"videos\": [\n {\n \"thumbnailUrl\": \"\",\n \"thumbnailDescription\": \"detailed visual description\",\n \"caption\": \"\",\n \"views\": <number>,\n \"format\": \"talking_head|skit|dance|tutorial|etc\",\n \"visualHooks\": []\n }\n ]\n },\n \"visualPatterns\": {\n \"filmingLocations\": [],\n \"lightingStyle\": \"\",\n \"thumbnailConsistency\": \"\",\n \"textOverlayUsage\": \"\",\n \"facePresenceRate\": \"\",\n \"productionQuality\": \"\"\n },\n \"contentMix\": {\n \"talkingHead\": \"%\",\n \"skits\": \"%\",\n \"trends\": \"%\",\n \"tutorials\": \"%\",\n \"other\": \"%\"\n },\n \"creatorArchetype\": \"\",\n \"brainScan\": {\n \"contentFormula\": \"\",\n \"trendChasingScore\": <0-100>,\n \"visualStyleAssessment\": \"\",\n \"thumbnailGameRating\": \"\",\n \"algorithmPrediction\": \"\",\n \"viralPotential\": \"\",\n \"improvements\": []\n },\n \"specificVideoExamples\": [\"<cite specific thumbnails>\"]\n}\n```\n\n---\n\n## TIKTOK VISUAL ANALYSIS NOTE\n\n**TikTok is thumbnail-driven. Analyze:**\n- The visual hook strategy\n- What makes someone stop scrolling\n- Patterns only visible by seeing the actual thumbnails\n- Reference specific videos in your assessment",
"userInput": "You are analyzing a TikTok profile with visual thumbnail analysis. Adapt to the UI.\n\n**TASK CONFIGURATION:**\n- Target: \n- Videos to Analyze: 15\n\n---\n\n## STEP 1: NAVIGATE TO PROFILE\n\n1. Parse target - extract username\n2. Navigate to tiktok.com/@[username]\n3. Wait for profile to fully load\n4. **Check accessibility:**\n - Profile exists? If not, report error\n - Private account? Report limitation\n - Videos visible? Proceed if yes\n\n---\n\n## STEP 2: EXTRACT PROFILE METADATA\n\n### Bio & Identity\n- Display name\n- Username\n- Bio text\n- Links (Linktree, etc.)\n- Verified status\n- Category/creator type label\n\n### Metrics\n- Follower count\n- Following count\n- Total likes\n- Video count\n\n### Profile Picture\n- Describe the profile picture\n- Style and vibe\n\n---\n\n## STEP 3: EXTRACT VIDEOS WITH THUMBNAILS\n\n**Extract 15 videos. For each capture:**\n\n1. **Thumbnail image URL** - the video preview/cover image\n2. **Video title/caption** - if visible\n3. **View count** - engagement indicator\n4. **Likes/comments** - if visible on hover\n5. **Duration** - video length\n6. **Sound/audio name** - if visible\n7. **Pinned status** - is this pinned?\n\n**Store structured:**\n```\nvideos: [\n {\n thumbnailUrl: \"url\",\n caption: \"text\",\n views: X,\n likes: Y,\n duration: \"0:15\",\n sound: \"sound name if visible\",\n isPinned: false\n }\n]\n```\n\n---\n\n## STEP 4: ANALYZE EACH THUMBNAIL (CRITICAL)\n\n**For each video thumbnail, generate detailed visual analysis:**\n\n### Describe:\n- **Scene/setting**: Where is this filmed? (bedroom, outdoor, studio, car, etc.)\n- **Person presence**: Face visible? Full body? Hands only? No person?\n- **Expression/pose**: If person visible - expression, energy, pose\n- **Text overlays**: Any text on thumbnail, what does it say?\n- **Visual hooks**: What makes this clickable? Curiosity gap?\n- **Lighting/quality**: Ring light, natural, low quality, professional\n- **Editing indicators**: Heavy effects visible in thumbnail?\n- **Format type**: Talking head, skit, dance, tutorial, reaction, duet, stitch\n- **Aesthetic**: Chaotic, polished, raw, curated\n- **Trend indicators**: Recognizable trend formats or templates\n\n### Pattern recognition across thumbnails:\n- Consistent visual style or random?\n- Face-forward thumbnails vs. not\n- Text-heavy vs. visual-only\n- Signature elements that repeat\n\n---\n\n## STEP 5: ANALYZE CONTENT PATTERNS\n\n### Video Format Mix\nBased on thumbnails, categorize:\n- Talking head %\n- Skits/comedy %\n- Dance/performance %\n- Tutorial/educational %\n- Reactions/duets %\n- Lifestyle/vlog %\n- Trend participation %\n\n### Visual Consistency\n- Same filming location recurring?\n- Consistent lighting setup?\n- Similar editing style?\n- Recognizable \"brand\" elements?\n\n### Thumbnail Strategy\n- Are thumbnails optimized for clicks?\n- Text overlay patterns\n- Expression patterns\n- Hook strategies visible\n\n---\n\n## STEP 6: AUDIO/TREND ANALYSIS\n\n**From visible sound names:**\n- Original sounds vs. trending sounds\n- Music genres favored\n- Trend participation rate\n- Sound originality score\n\n---\n\n## STEP 7: ENGAGEMENT PATTERN ANALYSIS\n\n**From view/like counts:**\n- Average performance\n- Viral hits (outliers)\n- Consistent performers\n- What visually differs in top performers?\n\n---\n\n## STEP 8: GENERATE BRAIN SCAN\n\n### 1. CREATOR ARCHETYPE\nBased on visual analysis, classify:\n- Examples: \"Reply Guy Energy\", \"Main Character\", \"Unhinged But Lovable\", \"Corporate TikTok Escapee\", \"Dance Trend Devotee\", \"Chaos Goblin\", \"Tutorial Queen\", \"Thirst Trap Artist\", \"Storytimer\", \"POV Actor\"\n- Must be based on ACTUAL visual patterns\n\n### 2. CONTENT FORMULA\n- What's their visual schtick?\n- Recurring thumbnail elements\n- Signature style markers\n\n### 3. TREND-CHASING SCORE (0-100)\nBased on visual analysis:\n- Original content indicators vs. trend templates\n- Format originality\n- Sound choices\n\n### 4. VISUAL STYLE ASSESSMENT\n- Filming quality and setup\n- Editing intensity\n- Thumbnail optimization level\n- Production value spectrum\n\n### 5. THUMBNAIL GAME RATING\n- How clickable are their covers?\n- Text overlay effectiveness\n- Expression optimization\n- Curiosity gap usage\n\n### 6. THE ALGORITHM KNOWS\nBased on visual patterns, predict:\n- What TikTok's algorithm thinks about them\n- Content categorization\n- Audience demographic guess\n\n### 7. VIRAL POTENTIAL ASSESSMENT\n- What visual elements are working\n- What's holding them back\n- Specific thumbnail/visual improvements\n\n---\n\n## STEP 9: REPORT RESULTS\n\n```json\n{\n \"target\": \"\",\n \"username\": \"\",\n \"profileMetrics\": {\n \"followers\": <number>,\n \"following\": <number>,\n \"totalLikes\": <number>,\n \"videoCount\": <number>\n },\n \"videosAnalyzed\": {\n \"count\": <number>,\n \"videos\": [\n {\n \"thumbnailUrl\": \"\",\n \"thumbnailDescription\": \"detailed visual description\",\n \"caption\": \"\",\n \"views\": <number>,\n \"format\": \"talking_head|skit|dance|tutorial|etc\",\n \"visualHooks\": []\n }\n ]\n },\n \"visualPatterns\": {\n \"filmingLocations\": [],\n \"lightingStyle\": \"\",\n \"thumbnailConsistency\": \"\",\n \"textOverlayUsage\": \"\",\n \"facePresenceRate\": \"\",\n \"productionQuality\": \"\"\n },\n \"contentMix\": {\n \"talkingHead\": \"%\",\n \"skits\": \"%\",\n \"trends\": \"%\",\n \"tutorials\": \"%\",\n \"other\": \"%\"\n },\n \"creatorArchetype\": \"\",\n \"brainScan\": {\n \"contentFormula\": \"\",\n \"trendChasingScore\": <0-100>,\n \"visualStyleAssessment\": \"\",\n \"thumbnailGameRating\": \"\",\n \"algorithmPrediction\": \"\",\n \"viralPotential\": \"\",\n \"improvements\": []\n },\n \"specificVideoExamples\": [\"<cite specific thumbnails>\"]\n}\n```\n\n---\n\n## TIKTOK VISUAL ANALYSIS NOTE\n\n**TikTok is thumbnail-driven. Analyze:**\n- The visual hook strategy\n- What makes someone stop scrolling\n- Patterns only visible by seeing the actual thumbnails\n- Reference specific videos in your assessment",
"urls": [],
"meta": {
"createdAt": "",
"userAgent": ""
},
"execution": {
"mode": "device",
"deviceIds": []
}
}
Add to Chrome