Sora API YouTube Shorts Uploader

Automated TikTok → YouTube Shorts Upload System Using n8n, Google APIs, and a Custom Frontend

This project is an end-to-end automation pipeline that:

1. Generates an AI video using OpenAI’s Sora API (or any video generation endpoint you connect)
2. Uploads the video to your Google Drive
3. Logs metadata to Google Sheets
4. Downloads the video for re-upload
5. Automatically uploads the video to your YouTube Shorts channel
6. Provides a custom frontend with buttons to trigger workflows & navigate to essential pages

---

🚀 How the Backend Works (n8n Workflow)

Below is a breakdown of every node exactly as they appear in the screenshot and what each one does.

🔴 1. Webhook

• Purpose: Starts the entire workflow when your frontend sends the prompt.
• Triggered by: Clicking the lightbulb → “Run Workflow” button on the frontend.
• Output: Receives the prompt text from your website.

---

🌐 2. HTTP Request (Sora Prompt)

• Purpose: Sends the user’s prompt to the OpenAI Sora API to start generating the video.
• Input: The prompt from the Webhook.
• Output: Returns a run_id from Sora.

---

🌐 3. HTTP Request1 (Polling Sora for Completion)

• Purpose: Every ~15 seconds, this node checks if Sora finished generating the video.

• Logic:

  • If status != "completed", loop back into itself (via the IF node).
  • When status == "completed", continue forward.

---

⏱️ 4. Wait

• Purpose: Creates a 15-second delay between each poll.
• Connected between: HTTP Request1 → Wait → back to HTTP Request1.

---

📥 5. HTTP Request2 (Download Sora Video URL)

• Purpose: After Sora finishes, this node fetches the actual file URL.

---

📤 6. Upload File (Google Drive)

• Requires:

  • Google OAuth2 credentials
  • Google Drive API enabled in Google Cloud Console

• Purpose: Uploads the downloaded Sora video to your Google Drive folder.

• Output: Produces a webViewLink (used later for Sheets + YouTube).

---

🔧 7. Code Node (Formatting Metadata)

Your current code (unchanged):

const soraData = $items("HTTP Request1")[0].json;

return [
  {
    json: {
      Title: soraData.prompt || soraData.input?.prompt || "No prompt found",
      Date: new Date().toLocaleString(),
      Video_Link: $json.webViewLink || $json.data?.webViewLink || "No link found"
    }
  }
];

Purpose:

• Pull prompt + timestamp
• Grab the Google Drive video link
• Format everything into a single object for Google Sheets

---

📄 8. Append Row in Sheet (Google Sheets)

• Requires:

  • Google Sheets API credentials

• Purpose: Writes one row to your log spreadsheet:

  • Prompt
  • Date
  • Video Drive link

---

⬇️ 9. Download File (Google Drive)

• Purpose: Downloads the uploaded video from Drive so it can be sent to YouTube.
• Input: The fileId from the previous Drive Upload node.

---

📺 10. Upload a video (YouTube)

• Requires:

  • YouTube Data API OAuth2 credentials
  • Your Google Cloud project must have: ✔ YouTube Data API v3 enabled ✔ OAuth client with “YouTube Data API” scopes

• Purpose: Uploads the downloaded video as a YouTube Short.

• Important:

  • You must use a valid YouTube category ID.
  • Shorts use categoryId = 22 (People & Blogs).
  • Title + description must come from earlier nodes.

• Example YouTube fields:

  • Title: Expression → {{$node["Code"].json["Title"]}}
  • Description: Check this out! #Shorts #AI #Viral #Short #Funny
  • Tags: Shorts,AI,Viral,Short,Funny
  • File: Binary → {{$binary.data}}

---

⚙️ Required API Integrations

1. Google OAuth2

Enable in Google Cloud:

• Google Drive API
• Google Sheets API
• YouTube Data API v3

Then create:

• OAuth Client ID (Desktop App)
• Paste credentials into n8n → Credentials → Google OAuth2

---

2. Google Sheets API

• Create a spreadsheet
• Share with your service account email
• Connect in n8n via Google Sheets node.

---

3. YouTube OAuth (YouTube Data API v3)

• Must enable YouTube Data API v3

• Scopes needed:

  • youtube.upload
  • youtube

• Select External or Internal OAuth consent screen

• Add your Google account as test user

---

🎨 Frontend (index.html UI)

The frontend is a lightweight static HTML page with a modern full-screen hero design.

The UI was designed from scratch (HTML + CSS + tiny bit of JS). It includes six circular icon buttons, each with a dedicated function.

---

🖥️ Buttons & What They Do

💡 Lightbulb Icon — Generate Prompt Ideas

• Calls a local JS prompt generator using ChatGPT
• Pops up a list of idea suggestions (You removed the OpenAI node from the backend to avoid costs — this stays front-end only.)

---

🏀 Basketball Icon — Trigger Workflow

• Sends the user’s prompt to your n8n webhook
• Starts the entire Sora → Drive → Sheets → YouTube pipeline

---

🐙 GitHub Icon — Open the Project Repo

• Links to: https://github.com/justinlej12/Sora-2-API-n8n-Automation

---

📷 Camera Icon — Open YouTube Channel

• Opens your YouTube channel where the Shorts are uploaded

---

📄 Sheets Icon — Open Organizer Sheet

• Links directly to your Google Sheets log
• Shows: Prompt, Date, Drive link, etc.

---

📁 Google Drive Icon — Open Video Folder

• Opens the Drive folder where n8n uploads each generated Sora video

---

🧩 How It All Fits Together

1. You open the frontend → click the Basketball button
2. It sends your prompt → n8n Webhook
3. n8n calls OpenAI Sora → gets run_id
4. n8n polls until the video is done
5. n8n downloads the video → uploads it to Google Drive
6. n8n logs metadata to Google Sheets
7. n8n uploads the video to YouTube Shorts
8. You can view results via the YouTube, Sheets, and Drive buttons

---

🎉 Result

Your workflow now:

• Requires zero manual uploading
• Posts Shorts automatically
• Tracks everything
• Gives you a clean frontend dashboard
• Has already gotten 2,400+ views and new subs on YouTube (W 💪)