Toxly
Sign in Create project

Public beta · Content safety API

Moderation decisions for products that move fast.

Classify user text, links, files, and Discord messages with structured decisions your backend can route.

Create a project Read the quickstart

API keys · Policies · Review queue · Discord guard

Product

A safety layer between user input and your app.

Toxly gives your backend a consistent first pass: classify the text, apply project policy thresholds, store a safe log, and return a decision your product can route.

Use it for chats, comments, profiles, prompts, support forms, listings, and Discord servers.

Text moderation API

One endpoint returns `allowed`, `decision`, `risk_score`, category scores, matched rules, and a short reason.

{
  "decision": "review",
  "risk_score": 0.63,
  "categories": {"sexual": 0.63}
}

Policy engine

Thresholds convert scores into product actions. Keep strict rules for production and softer rules for testing.

toxicity > 0.75  block
self_harm > 0.65  escalate
pii      > 0.80  mask

Moderation logs

Debug decisions without storing the full original text. Logs keep a hash, preview, score, decision, categories, and latency.

mod_2fTZ...block0.82 mod_PvtK...review0.63 mod_Hf5...mask0.86
Workflow

Request to action.

  1. 01

    Request

    Your backend sends user-generated text to Toxly.

  2. 02

    Category scores

    The response includes scores from 0.0 to 1.0 for each configured safety category.

  3. 03

    Policy thresholds

    Project rules decide whether the content should pass, wait for review, be masked, or be blocked.

  4. 04

    Application action

    Your product routes `allow`, `warn`, `review`, `block`, `mask`, or `escalate`.

Dashboard

Projects, keys, policies, logs, and usage.

The dashboard stays close to the API surface. It shows the parts developers need to operate moderation in production.

Mathingo

Create API key

Requests today1,284

Block rate4.8%

Avg latency220 ms

RequestDecisionRiskLatency mod_2fTZ...block0.82220 ms mod_PvtK...review0.63241 ms mod_Hf5...mask0.86198 ms
Categories

Scores you can inspect.

The response keeps category scores visible so teams can tune policies without guessing.

toxicityhateharassmentviolence self_harmsexualminor_safetyspam scampiijailbreakprompt_injection
Pricing

Simple limits for early teams.

PlanPriceRequests per month
Free0 EUR1,000
Starter9 EUR50,000
Pro29 EUR250,000
Scale79 EUR1,000,000
EnterpriseCustomby agreement
Public beta

Currently in public beta.

The core text moderation API, project policies, logs, usage limits, members, support tickets, and Discord setup are available. Image moderation and deeper model routing are planned next.

Built for pilots

Projects, members, API keys, policies, usage, support tickets, billing, and Discord setup are ready for early company testing.

Security baseline

Hashed API keys, CSRF protection, rate limits, request limits, Turnstile support, safe logs, and separate API/dashboard/docs hosts.

Operational docs

Quickstart, API reference, errors, policies, rate limits, Discord setup, legal pages, and production readiness notes.

Send your first moderation request.

Create a project, generate an API key, and follow the quickstart from your backend.

Create a project Open documentation