Tutorials, research and playbooks on AI content moderation. Built by the team behind ToxicFilter, written for developers and trust & safety teams.
Build a comment wall in Express that publishes, holds or refuses each comment with its reason, keeps the key on the server, and verifies the review webhook.
Build a comment wall in Flask that publishes, holds or refuses each comment with its reason, verifies the review webhook, and runs its tests without the network.
A validation rule, a facade and a fake: build a comment wall in Laravel that publishes, holds or refuses each comment, and publishes held ones when a person approves them.
No framework: a comment wall in plain PHP that publishes, holds or refuses each comment, tells the author why, and publishes held ones when a person approves them.
Run a thin moderation proxy at the edge. Lower latency, smaller blast radius on outages, and a simpler client-side integration.
Exponential backoff, jitter, circuit breakers, and what to do when the API is briefly unavailable. Keep your app resilient.
Moderating after upload means the unsafe content already exists on your infrastructure. Here is how to flip the order of operations.
Blocking the user until moderation completes is not always the right answer. A decision framework for sync, async, and hybrid patterns.
Allow, review or block, and the reason in words. On the free plan: 2,000 credits a month, no card. A check costs 1 credit, about 8 if the model reads it, about 10 for an image.
