AI Social Campaign Generator

This project explores an automated workflow for generating complete social media campaigns using AI. A Python script sends structured prompts to Claude to produce platform-specific marketing copy, while Stable Diffusion generates supporting imagery. The goal was to test how AI tools can be orchestrated into a repeatable pipeline that produces both strategy and visual assets for a brand.

What This Solves

Creating consistent, multi-platform social campaigns is time-consuming and often fragmented across copywriting and visual production.

This system demonstrates how AI tools can be orchestrated into a single workflow that generates both messaging and visual direction, reducing turnaround time and improving campaign consistency.

The Problem

Creating a complete social media campaign normally requires both marketing copy and visual assets. These tasks are often produced separately and manually, making the process slow and inconsistent. This experiment explored whether a single automated workflow could generate both strategy and imagery from structured prompts.

The Approach

To test this idea, a Python script was developed to orchestrate multiple AI tools into a single workflow. The script sends structured prompts to Claude to generate marketing strategy and platform-specific post copy. It then constructs detailed prompts for Stable Diffusion to generate supporting imagery aligned with the campaign theme. By combining these steps into one automated pipeline, the system produces both marketing text and visual assets from a single input.

Tools Used

  • Python
  • Claude API
  • Stable Diffusion
  • Prompt Engineering
  • Automated Workflow Design

Workflow

  • User inputs a brand name and product.
  • A Python script constructs a structured prompt and sends it to the Claude API.
  • Claude generates a marketing strategy and platform-specific social media posts.
  • The script extracts key themes from the response and constructs image prompts.
  • These prompts are sent to Stable Diffusion to generate supporting campaign visuals.
  • The resulting text and images form a complete social media campaign package.

Example Output

The system generates a complete campaign package including platform-specific copy, hashtags, and image prompts suitable for AI image generation. Below is an example campaign produced by the workflow.

NovaSkin_campaign_2026-03-04Download

Key Takeaways

  • AI tools can be orchestrated into a structured pipeline rather than used as isolated generators.
  • Prompt engineering plays a critical role in shaping both marketing strategy and visual generation outputs.
  • Combining language models and generative image models enables rapid prototyping of full campaign concepts.