Automating AI-Powered Instagram Reels with Make.com, OpenAI & Google Cloud

Create, voiceover, and publish stunning Instagram Reels — automatically — using only a text prompt.

Overview

This project presents a fully automated pipeline that transforms a text prompt into a narrated Instagram Reel video using:

• Make.com to orchestrate the automation
• OpenAI to generate text, audio, and images
• Google Drive to store backup files
• Google Cloud Storage (GCS) as hosting for Cloud Run
• Google Cloud Run to merge the PNG image and MP3 audio into an MP4 video

With a pre-built scenario and Docker-based Cloud Run service, this system takes care of everything: generating content, creating media, uploading it, and optionally publishing to Instagram.

Complete Guide: Automating AI-Powered Instagram Reels with Make.com, OpenAI, and Google Cloud

Introduction

This detailed guide walks you step-by-step through setting up and launching a pre-configured Make.com workflow that automatically generates and publishes Instagram Reels using just a text prompt. The workflow is provided as a JSON file and only needs to be imported into Make.com, then configured properly.

---

🧭 How to Use Make.com (Beginner Guide)

What is Make.com?

Make.com is a no-code automation platform similar to Zapier, but far more flexible and powerful. It allows you to build automation flows (scenarios) composed of connected modules. Each module performs a task: uploading a file, sending an API request, processing data, and so on.

Key Concepts:

• Scenario: the automation flow
• Module: a building block that performs an action (e.g., generate audio, upload to Drive)
• Variable/Output: the result from a module that can be used in another one (drag & drop)

Importing the Workflow JSON

1. Go to make.com
2. Log in (or sign up)
3. Navigate to "Scenarios"
4. Click "Import scenario"
5. Upload the .json file provided (e.g., workflow-ai-reels.json)
6. Open the imported scenario — all modules will be pre-connected
7. Proceed with configuring each module (authentication and field values)

Connecting Accounts (Authentication)

For each module (e.g., OpenAI, Google Drive), click "Add" or "Connect" and:

• Enter your API key (for OpenAI)
• Provide OAuth credentials (for Google)
• Authorize account access (a pop-up window will appear)

If a module doesn’t show output, click “Run once” to test it manually.

---

🧱 How to Configure Google Cloud Platform (GCP) — Step-by-Step

1. Create a GCP Project

1. Go to console.cloud.google.com
2. Click "Project" (top left) > "New Project"
3. Name your project (e.g., makecom-reels)
4. Save and wait for it to finish creating

2. Enable Required APIs

Go to:

• Menu > "APIs & Services" > "Library"
  • Search and enable:
    • Cloud Resource Manager API
    • Cloud Pub/Sub API
    • Cloud Run API
    • Google Drive API
    • Cloud Storage API

3. Create a Google Cloud Storage Bucket

1. Menu > "Storage" > "Browser"
2. Click “Create Bucket”
3. Name it: makecom-reels
4. Region: Multi-region – Europe
5. Access: Make public (optional but helpful)
6. Click Create

4. Create an OAuth Client (Google Drive + GCS)

1. Menu > APIs & Services > OAuth Consent Screen

   • User type: External
   • App name: makecom-pr-client
   • Developer email: your own
   • Authorized domains:
     • https://make.com
     • https://integromat.com
     • https://www.googleapis.com
   • Save

2. Menu > Credentials > "Create credentials" > "OAuth client ID"

   • Application type: Web App
   • Name: makecom-pr-client
   • Redirect URIs:
     • https://www.integromat.com/oauth/cb/google-custom
   • Save and copy Client ID and Client Secret

3. Paste your Client ID/Secret into Make.com when configuring Google Drive or GCS modules.

---

🔁 Full Workflow – Step-by-Step Instructions

This section explains each automation step from the pre-configured Make.com scenario. Each step corresponds to a module in the workflow.

Step 1: Generate Text with OpenAI (GPT-4o)

• Module: OpenAI – Chat Completion
• Model: gpt-4o
• Purpose: Generate a short, spoken-style script for a Reel
• Prompt example:

  Write a short, engaging script for an Instagram Reel about X
  

Step 2: Convert Text to Voice (TTS)

• Module: OpenAI – Generate Audio
• Model: tts-1
• Input: Output from Step 1
• Output: MP3 voiceover

Step 3: Upload MP3 to Google Drive

• Module: Google Drive – Upload File
• Purpose: Backup the MP3 file
• OAuth: Set up via Google Developer Console with redirect URIs
• Folder: /MakeCom/AI-reels/

Step 4: Upload MP3 to GCP Bucket

• Module: HTTP – Upload to Google Cloud Storage (via signed URL or Make.com integration)
• Bucket: makecom-reels

Step 5: Generate Image with DALL·E 3

• Module: OpenAI – Generate Image
• Prompt example:

  A modern, eye-catching illustration about CyberSeXcurity for a vertical Reel cover.
  Bright, clean, minimalistic – no text or logos.
  

• Size: 1024x1792 px
• Style: Vivid

Step 6: Get the Image File (URL)

• Module: HTTP – Get a File
• URL: Use {{3.data[].url}} from the DALL·E response

Step 7: Upload PNG to Google Drive

• Module: Google Drive – Upload File
• Purpose: Backup of the generated image

Step 8: Upload PNG to GCP Bucket

• Module: HTTP – Upload (or GCP module)

Step 9: Merge Audio + Image with Cloud Run

• Service: Google Cloud Run (custom Docker image)
• Tools: ffmpeg, Flask, gunicorn
• Required files: main.py, Dockerfile, requirements.txt

gcloud builds submit --tag <docker-image-url>
gcloud run deploy video-editor --image <docker-image-url> --platform=managed --region=europe-west3 --allow-unauthenticated

Test endpoint:

curl -X POST "https://video-editor-xxxxxx.run.app/process"   -H "Content-Type: application/json"   -d '{"image":"gs://makecom-reels/file.png","audio":"gs://makecom-reels/audio.mp3","output":"gs://makecom-reels/output.mp4"}'

Step 10: Retrieve the Generated Video File

• Module: HTTP – Make a Request
• API: Google Cloud Storage List Objects
• URL: https://www.googleapis.com/storage/v1/b/makecom-reels/o?orderBy=timeCreated
• Query:
  • name: prefix
  • value: output-

Step 11: Upload MP4 to Google Drive

• Module: Google Drive – Upload File
• Purpose: Backup and easy access to video

Step 12: Optional – Publish to Instagram (Manual or via API)

Post manually or use third-party schedulers.

Step 13: Cleanup Temporary Files

• Module: HTTP or GCP – Delete Object
• Files: file.png, openai-tts-output.mp3, output.mp4
• Add error handling: "Ignore if file doesn’t exist"

Step 14: Send Email Notification (Optional)

• Service: MailerSend SMTP or HTTP API
• Domain: trial-xxxxxx.mlsender.net

---