Catfish Wrangler checks new comments, reviews, and uploaded images against the Content Guard API and flags anything suspicious with a color-coded badge in your WordPress admin — right in the Comments list and the Media Library, alongside content you already review.

What it checks for:

Classic romance-scam language in comments and profile-style text (e.g. “stationed overseas,” refusal to video call, requests to move off-platform or send money)
Generic template-style fake review language
Whether an uploaded image looks AI-generated (via a live visual AI-detection model)
Whether an image is missing the camera metadata a real photo would normally have
Whether a profile’s photos are inconsistent with each other (some carrying real camera metadata, others not)

Safe by default. Out of the box, Catfish Wrangler only flags — it never holds or blocks a comment automatically. You review the flagged items yourself. Once you’ve seen it work correctly on your own site, you can turn on auto-hold (with a configurable sensitivity threshold) from Settings > Catfish Wrangler.

Requires an API key. Catfish Wrangler analyzes content using the Content Guard API, a paid, metered service — this plugin does not work without an account and an API key, which you enter yourself under Settings > Catfish Wrangler. Nothing is sent anywhere until you configure a key.

Why this matters

Dating profiles, marketplace listings, and review sections increasingly attract AI-generated photos and scripted scam text. This plugin gives site owners and moderators a heads-up on suspicious content without having to manually cross-check every submission.

External Services

This plugin connects to the Content Guard API to analyze comment/review text and uploaded images for signs of fraud or AI-generation.

What is sent: the text of new comments/reviews, and newly uploaded image files, are sent to the Content Guard API for analysis.
When: only when this plugin is active and configured with an API key — every new comment/review and every new image upload is sent for analysis at the time it’s submitted.
Why: to compute the suspicion signals (scam-language patterns, AI-generation likelihood, metadata checks) shown as a badge in your admin area.
Service URL: the API endpoint is configurable under Settings > Catfish Wrangler (defaults to https://content-guard-production-24a7.up.railway.app).
Provider terms: Terms of Use / Privacy & Data Handling — this also discloses that image analysis is partly performed by a sub-processor, Hive Moderation (thehive.ai), under their own terms.

No data is sent to the Content Guard API unless you have entered an API key in the plugin settings — this is an explicit opt-in step.