Complete request examples for all three modes. Copy, paste, and customize for your use case.
All examples use the /api/demo server route which proxies to https://api.sentienceapi.com/v1/observe
Extract clean, structured content from any webpage. Perfect for content extraction, RAG, and text analysis.
curl -X POST https://api.sentienceapi.com/v1/observe \
-H "Authorization: Bearer sk_live_..." \
-H "Content-Type: application/json" \
-d '{
"url": "https://www.shopify.com/pricing",
"mode": "read",
"format": "markdown",
"options": {
"viewport": {
"width": 1920,
"height": 1080
},
"contentLimit": 50000
}
}'Get precise coordinates and geometry data for all interactable elements. Ideal for automation, RPA, and element detection.
curl -X POST https://api.sentienceapi.com/v1/observe \
-H "Authorization: Bearer sk_live_..." \
-H "Content-Type: application/json" \
-d '{
"url": "https://github.com/facebook/react",
"mode": "map",
"options": {
"viewport": {
"width": 1920,
"height": 1080
},
"limit": 100,
"filter": {
"allowed_roles": ["button", "link", "textbox"],
"min_area": 100,
"min_z_index": 0
},
"render_quality": "performance",
"include_visual_cues": false
}
}'Capture screenshots with bounding box overlays. Perfect for visual analysis, e-commerce scraping, and screenshot-based AI training.
curl -X POST https://api.sentienceapi.com/v1/observe \
-H "Authorization: Bearer sk_live_..." \
-H "Content-Type: application/json" \
-d '{
"url": "https://www.shopify.com/pricing",
"mode": "visual",
"options": {
"viewport": {
"width": 1920,
"height": 1080
},
"limit": 50,
"render_quality": "precision"
}
}'Get visual styling hints (color names, cursor type, prominence) for icon-heavy UIs and design tools. Perfect for recognizing icon-only buttons.
curl -X POST https://api.sentienceapi.com/v1/observe \
-H "Authorization: Bearer sk_live_..." \
-H "Content-Type: application/json" \
-d '{
"url": "https://figma.com/design/123",
"mode": "map",
"options": {
"viewport": {
"width": 1920,
"height": 1080
},
"limit": 200,
"include_visual_cues": true
}
}'💡 Response includes: Each element will have is_occluded (always) and visual_cues object with background_color_name, color_name, cursor, and is_primary fields.