Feed not processing properly | CyberSEO Pro | Support Forum

I've checked your feed settings and I'm confused about what exactly you're trying to do there. You are using the outdated "AI article generation" method (not recommended, but still a working solution), and at the same time you are using the [openai_gpt] shortcode. In fact, you should use one of these methods, but not both together.

You first rewrite the article with one prompt and then change the result with another. This is a waste of time and the money you pay for the OpenAI GPT API. All this can be done with a single prompt using the appropriate shortcode.

Also you can't do something like this:

Login to see the code

The OpenAI GPT API does not have access to the Internet, so it can't analyze online content via a link.

You should disable "AI article generation" settings and use only shortcodes like this:

Login to see the code

If you want to rewrite the title, use a template shortcode for that as well.

However, I have not been able to reproduce your problem and the post has been successfully generated, so I suggest you to analyze your server error log. Perhaps the script fails due to an error caused by misconfiguration of your server, for example, some required PHP extension may be missing in your config.

Login to see the quote

As for AI, there's a learning curve for all of us, since it's such a new technology. Remember, process optimization is just a classic programmer habit, and you're already making progress by tweaking your approach. Noticing that you were running the OpenAI GPT API twice on the same text was a key observation. Ideally, you should streamline your process by using a single, well-designed prompt that can handle both summarization and conclusion. This method not only saves resources, but also ensures consistency of the generated content.

Also, consider using the %post_content_notags% placeholder instead of %post_content%. This small change passes the article content to the GPT without any HTML tags, further reducing the amount of data processed and thus your API usage costs. Keep experimenting and tweaking-you're definitely on the right track, and things should run more smoothly now.