Generate social posts and images with Claude, Gemini, and OpenAI

Description

Quick overview
This workflow receives a webhook request with a topic and context, generates platform-specific copy for LinkedIn, X, Facebook, and Instagram using Anthropic Claude, Google Gemini, or OpenAI, optionally creates a matching image via OpenAI Images or Gemini Flash Image, and returns the results as JSON.

How it works
Receives a POST request on the /webhook/social-content webhook containing the topic, context, provider selection, and optional tone/audience/brand/CTA/language and image settings.
Normalizes and validates the payload (including provider, content length limits, and image options) and returns a 400 response if required fields are missing or invalid.
Routes the request to Anthropic Claude, Google Gemini, or OpenAI to generate structured social posts for LinkedIn, X (single post and thread), Facebook, and Instagram.
If image generation is enabled and requested, routes to OpenAI Images (gpt-image-1) or the Gemini image generation API to create an image prompt based on the Instagram image idea and requested style/size.
Merges the generated text and optional image (or an image error) into a single JSON payload.
Responds to the webhook caller with a 200 success JSON payload, or a 500 error payload if text generation fails.

Setup
Add credentials for at least one text provider you plan to use: Anthropic API (Claude), Google Gemini / PaLM API, and/or OpenAI API.
If you want image generation, also configure the required credential for the chosen image provider (OpenAI API for gpt-image-1 and/or Google Gemini / PaLM API for Gemini image generation).
Set your preferred default behavior in the workflow settings payload (for example, toggle the image_generation_enabled master switch) and adjust the prompts/models/temperatures in the respective provider nodes as needed.
Activate the workflow and copy the production webhook URL into your calling app or script that will send POST requests to /webhook/social-content.