Quickstart

Prerequisites

A ScaleReach account with an active plan (minutes balance > 0)
An API key (get one here)

Step 1: Submit a YouTube video

With an API key, the workspace is auto-resolved — you don't need to fetch it separately.

curl -X POST https://api.scalereach.ai/api/videos/youtube \
  -H "Authorization: Bearer sr_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "youtubeUrl": "https://www.youtube.com/watch?v=VIDEO_ID",
    "config": {
      "clipType": "viral-clips",
      "clipDurationMin": 15,
      "clipDurationMax": 90,
      "aspectRatio": "9:16",
      "enableCaptions": true,
      "enableAutoHook": true,
      "genre": "Auto",
      "clipModel": "ClipBasic"
    }
  }'

Minutes are deducted immediately based on video duration. A 30-minute video costs ~30 minutes.

Save the video.id from the response.

Step 2: Poll for completion

curl https://api.scalereach.ai/api/videos/VIDEO_ID/status \
  -H "Authorization: Bearer sr_live_YOUR_KEY"

The video goes through these stages:

Status	Description
`downloading`	Downloading from YouTube
`uploading`	Uploading to cloud storage
`transcribing`	AI speech-to-text
`analyzing`	AI detecting viral moments
`generating_clips`	Cutting and rendering clips
`completed`	All clips ready
`failed`	Something went wrong (check `errorMessage`)

Poll every 10-15 seconds. A typical 30-minute video takes 3-5 minutes to fully process.

Step 3: Get your clips

Once status is completed:

curl https://api.scalereach.ai/api/videos/VIDEO_ID/clips \
  -H "Authorization: Bearer sr_live_YOUR_KEY"

Response:

{
  "videoId": "VIDEO_ID",
  "clips": [
    {
      "id": "clip_abc123",
      "title": "The moment everything changed",
      "startTime": 245,
      "endTime": 302,
      "duration": 57,
      "score": 92,
      "viralityReason": "Strong emotional hook with unexpected twist",
      "transcript": "And that's when I realized...",
      "downloadUrl": "https://cdn.scalereach.ai/signed-url...",
      "thumbnailUrl": "https://cdn.scalereach.ai/thumb...",
      "status": "completed",
      "aspectRatio": "9:16"
    }
  ],
  "count": 8
}

Each clip has a downloadUrl (signed, expires in 1 hour) you can download directly.

Full script

#!/bin/bash
API_KEY="sr_live_YOUR_KEY"
BASE="https://api.scalereach.ai"

# 1. Get workspace ID
WORKSPACE_ID=$(curl -s "$BASE/api/workspaces" \
  -H "Authorization: Bearer $API_KEY" \
  | python3 -c "import sys,json; print(json.load(sys.stdin)[0]['id'])")
echo "Workspace: $WORKSPACE_ID"

# 2. Submit video
VIDEO_ID=$(curl -s -X POST "$BASE/api/videos/youtube" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d "{
    \"youtubeUrl\": \"https://www.youtube.com/watch?v=VIDEO_ID\",
    \"workspaceId\": \"$WORKSPACE_ID\",
    \"config\": {
      \"clipType\": \"viral-clips\",
      \"aspectRatio\": \"9:16\",
      \"enableCaptions\": true
    }
  }" | python3 -c "import sys,json; print(json.load(sys.stdin)['video']['id'])")
echo "Video ID: $VIDEO_ID"

# 3. Poll until completed
while true; do
  STATUS=$(curl -s "$BASE/api/videos/$VIDEO_ID/status" \
    -H "Authorization: Bearer $API_KEY" \
    | python3 -c "import sys,json; print(json.load(sys.stdin)['video']['status'])")
  echo "Status: $STATUS"
  if [ "$STATUS" = "completed" ] || [ "$STATUS" = "failed" ]; then break; fi
  sleep 15
done

# 4. Get clips
curl -s "$BASE/api/videos/$VIDEO_ID/clips" \
  -H "Authorization: Bearer $API_KEY" | python3 -m json.tool

Python example

import requests
import time

API_KEY = "sr_live_YOUR_KEY"
BASE = "https://api.scalereach.ai"
HEADERS = {"Authorization": f"Bearer {API_KEY}"}

# 1. Get workspace
workspaces = requests.get(f"{BASE}/api/workspaces", headers=HEADERS).json()
workspace_id = workspaces[0]["id"]

# 2. Submit video
resp = requests.post(f"{BASE}/api/videos/youtube", headers=HEADERS, json={
    "youtubeUrl": "https://www.youtube.com/watch?v=VIDEO_ID",
    "workspaceId": workspace_id,
    "config": {"clipType": "viral-clips", "aspectRatio": "9:16", "enableCaptions": True},
})
video_id = resp.json()["video"]["id"]
print(f"Video submitted: {video_id}")

# 3. Poll
while True:
    status = requests.get(f"{BASE}/api/videos/{video_id}/status", headers=HEADERS).json()
    s = status["video"]["status"]
    print(f"Status: {s} ({status.get('job', {}).get('progress', 0)}%)")
    if s in ("completed", "failed"):
        break
    time.sleep(15)

# 4. Get clips
clips = requests.get(f"{BASE}/api/videos/{video_id}/clips", headers=HEADERS).json()
for clip in clips["clips"]:
    print(f"[{clip['score']}] {clip['title']} — {clip['downloadUrl']}")