🧠 Boost Your Geeklog SEO with AI-Generated Meta Tags

Automatically create meta descriptions, keywords, and SEO titles for your articles

In modern SEO, small details often make a big difference. Meta tags such as meta_description, meta_keywords, and page_title still play an essential role in how your content is displayed and understood by search engines. They directly influence how your stories appear in results, and therefore how often users click them.

I’ve recently built a simple, lightweight utility that helps Geeklog site administrators automatically generate these fields using OpenAI’s API. It works with any Geeklog 2.x site and doesn’t require additional plugins or modifications.

---

⚙️ Simple, fast, and content-friendly automation

The Geeklog AI SEO Updater connects directly to the Geeklog database and detects any stories missing a meta_description. For each one, it uses the gpt-4o-mini model to analyze the article’s title and introduction, then generates:

• a short optimized meta description (≤ 155 characters),
• a list of relevant meta keywords (5–10 per story),
• and a SEO-friendly page title (≤ 65 characters).

The script automatically detects the language of your site from $_CONF['language'] — it works in French, English, German, Spanish, Italian, and Japanese. You can test results safely in preview mode (?dry=1) or apply them in update mode with a defined article limit.

---

🔍 Why meta tags still matter

While modern search engines rely on full-page analysis, meta descriptions remain key to how your site looks in results. They act as your article’s “headline” on Google or Bing and have a strong effect on click-through rate.

Meta keywords, even though no longer used by Google for ranking, remain valuable inside Geeklog. They can be used for internal linking or automated features such as:

• a related stories system that connects articles with similar topics,
• a tag cloud for quick exploration,
• or a contextual search module based on shared keywords.

💡 For example, another script could automatically link stories that share two or more identical keywords, creating a stronger internal SEO structure.

---

🚀 Benefits for site administrators

• Save hours updating metadata across hundreds of stories.
• Keep consistent SEO formatting site-wide.
• Improve visibility and click performance in search results.
• Optionally notify Bing through the IndexNow plugin after each update.

The script only updates stories with empty meta descriptions and checks that you have the proper story.edit admin rights before execution.

---

🧩 Example usage

# Preview mode (no database update)
https://yourdomain.com/admin/utils/update_meta.php?dry=1

# Update 10 stories
https://yourdomain.com/admin/utils/update_meta.php?limit=10

You can also schedule it as a weekly cron job to automatically optimize metadata for newly published stories.

---

💡 A foundation for smarter features

This script lays the groundwork for more semantic features inside Geeklog. The generated keywords can later feed new modules such as:

• AI-related stories based on keyword similarity,
• smart internal linking or tag clusters,
• or even advanced search filtering.

By combining Geeklog’s simplicity with AI-assisted metadata, content becomes easier to navigate and better structured for both users and search engines.

---

📦 Availability

The full documented code is available on the Geeklog Community Forum: 👉 https://www.geeklog.net/forum/index.php?forum=10

It supports PHP 5.6 +, Geeklog 2.x, and optionally integrates with the IndexNow plugin to notify Bing. Every line of the shared version is commented for easy understanding and adaptation.

---

🧭 Final thoughts

With this AI-powered updater, Geeklog proves it can stay modern and efficient without losing its core simplicity. This approach helps keep older content relevant, visible, and better connected — with minimal effort and full control from the admin panel.

🧩 Author: Ben, Geeklog community contributor Discuss and download on the forum: https://www.geeklog.net/forum/index.php?forum=10