drupal/ai_recipe_comment_spam

Adds AI-powered spam scoring, reasoning and auto-unpublishing for comments, with a moderation triage view.

1.0.0-alpha1 2026-07-08 10:23 UTC

Requires

• drupal/ai: ^1.4
• drupal/core: ^11.2
• drupal/custom_field: ^4.0
• drupal/token: ^1.15

GPL-2.0-or-later

Published on Packagist.org on 2026-07-08 08:38 UTC

README

What it does

AI Comment Spam Moderation adds an AI-assisted workflow for comments. It scores each incoming comment for spam risk, saves the score and a one-sentence reasoning, automatically unpublishes likely spam, and surfaces everything in a moderation triage view.

Why it matters

Site moderators can triage comment spam faster and reduce manual review effort. Obvious spam is unpublished automatically, while borderline cases stay visible with clear, AI-generated reasoning for a human to act on.

How it works

When applied, the recipe configures:

• core comment support using the default comment comment type (via the core article_comment recipe),
• a Custom Field on comments (field_spam_assessment) holding a spam_score (1-10) and a spam_reasoning sentence,
• an AI automator on field_spam_assessment that reads the comment body and writes the score and reasoning (reasoning is the literal ok when the score is below 5),
• a second AI automator that unpublishes the comment when the spam_score is 8 or higher,
• a dedicated listing at /admin/content/ai-comment-spam for moderation triage.

Requirements & compatibility

• Drupal core: ^11.2
• AI module: ^1.4
• Custom Field: ^4.0
• Token: ^1.15

How to get started

1. Install module dependencies (including AI and AI Automators).
2. Configure an AI provider/model and set a default for chat_with_complex_json.
3. Apply the recipe.
4. Post comments on an Article and verify the spam score and reasoning are saved, and that high-scoring comments are unpublished.
5. Use /admin/content/ai-comment-spam for moderation triage.