Device-Analyze API — Detekcia User-Agent | YEB

#What Device-Analyze solves

This endpoint parses a User-Agent (UA) string and returns browser, engine, OS, device and bot signals, plus convenient booleans (is_mobile, is_desktop, …). Use it to tailor UX (mobile vs desktop layouts), segment analytics, or flag suspicious UAs.

#Endpoints & when to use them

# POST /v1/device-analyze

• Best for: All UA parsing use-cases. Single, simple endpoint.
• How it works: Provide a ua string (optional). If omitted, we read the request header.
• Common uses: Feature flags (e.g., disable heavy effects on mobile), funnel analysis, anti-fraud heuristics.

#Quick start

curl -X POST "https://api.yeb.to/v1/device-analyze" \
  -H "Accept: application/json" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: <YOUR_API_KEY>" \
  -d '{
    "api_key": "<YOUR_API_KEY>",
    "ua": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/141.0.0.0 Safari/537.36"
  }'

// JS Fetch example (Node/Browser)
await fetch("https://api.yeb.to/v1/device-analyze", {
  method: "POST",
  headers: {
    "Accept": "application/json",
    "Content-Type": "application/json",
    "X-API-Key": "<YOUR_API_KEY>"
  },
  body: JSON.stringify({
    api_key: "<YOUR_API_KEY>",
    ua: navigator.userAgent // or a server-provided UA
  })
}).then(r => r.json()).then(console.log);

#Parameters that actually matter

#Reading & acting on responses

You typically read data.browser, data.os, data.device, and boolean flags like is_mobile.

{
  "data": {
    "ua_string": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) ... Chrome/141.0.0.0 Safari/537.36",
    "browser": { "name": "Chrome", "version": "141.0.0.0", "vendor": null },
    "engine":  { "name": "Blink",  "version": null },
    "os":      { "name": "Windows","version": "10.0" },
    "device":  { "type": "desktop","brand": null,"model": null },
    "bot": null,
    "is_mobile": false,
    "is_tablet": false,
    "is_bot": false,
    "is_desktop": true
  },
  "response_code": 200,
  "response_time_ms": 118
}

#Key signals & decisions

• is_mobile / is_tablet / is_desktop — pick layout, image sizes, feature flags.
• bot object — treat as crawler: skip animations, block login, different rate limits.
• browser.version — gate advanced features (e.g., WebGPU) behind minimum versions.
• device.type — “mobile”, “tablet”, “desktop”, etc. for coarse segmentation.

#Errors you might see

#Practical recipes

• Mobile-first UI: if is_mobile → reduce bundle, enable compact nav, lazy-load heavy widgets.
• Fraud hardening: if is_bot or UA looks impossible (e.g., iOS + Edge legacy) → challenge or block.
• Analytics buckets: group by device.type and os.name for clean dashboards.

#Troubleshooting & field notes

1. Empty vendor/version: some UA strings are intentionally reduced. Don’t fail hard; fall back to booleans.
2. Proxy chains: ensure your edge forwards User-Agent unchanged if you rely on header fallback.
3. Batch analysis: always pass ua explicitly to avoid environment/header differences.

#More response examples

{
  "data": {
    "ua_string": "Mozilla/5.0 (Linux; Android 10; K) ... Chrome/138.0.0.0 Mobile Safari/537.36",
    "browser": { "name": "Chrome", "version": "138.0.0.0", "vendor": null },
    "engine":  { "name": "Blink", "version": null },
    "os":      { "name": "Android", "version": "10" },
    "device":  { "type": "mobile", "brand": null, "model": null },
    "bot": null,
    "is_mobile": true,
    "is_tablet": false,
    "is_bot": false,
    "is_desktop": false
  }
}

#API Changelog