You can now create custom moderation rules for every API key directly from your dashboard.
With Moderation Rules, you decide exactly how ModAPI should handle moderation scores and what additional data should be returned to your application.
What You Can Do Custom Flagging Thresholds Override the default 0.70 flagging threshold with your own logic.
Examples: • Flag content when the scam score is ≥ 0.50 • Don't flag content unless harassment is ≥ 0.90 • Apply different thresholds to text, image, or scam moderation
Custom Response Annotations Automatically inject your own JSON into API responses when conditions match.
Example: { "annotations": { "tier": "gold", "review_required": true } }
This makes it easy to route content, trigger workflows, assign review queues, or add metadata without maintaining extra logic on your backend.
New Response Field: Hash All moderation endpoints now return a deterministic SHA-256 request hash: { "flagged": false, "confidence": { ... }, "hash": "f156fb89...", "cached": false, "latency_ms": 359 }
The hash can be used for debugging, audit trails, cache verification, and identifying repeated content.
Rules Are Applied Per API Key Each API key can have its own independent moderation behavior, allowing different applications and environments to use different moderation policies.
Create and manage your rules now from https://modapi.xyz/dashboard/rules.
Build moderation workflows that fit your application instead of forcing your application to fit a single threshold.