Perspective API Alternative (Toxicity Detection)

io.github.koyourmoonv1.0.9Updated Oct 10, 2026

Toxicity scores (0-1) for text before an agent posts it. Perspective API compatible. No LLM.

VerifiedStreamable HTTPWeb executableAI & MLSecurity & Monitoring

Overview

AI-generated overview

Scores text for toxicity before an assistant posts it, returning six 0-1 scores and a flagged value.

What it does
This remote MCP server gives an assistant a tool to check text before publishing it. For each text it returns six scores from 0 to 1 plus a flagged value: TOXICITY, SEVERE_TOXICITY, INSULT, PROFANITY, THREAT and IDENTITY_ATTACK. It uses the open-source Detoxify model, not an LLM, so the same text always gets the same score. It is described as a drop-in replacement for the Google Perspective API and is English only.
When to use it
Use it when an assistant drafts or sends user-facing text and you want a deterministic moderation check before it goes out, for example comments, replies or generated messages. It also suits teams replacing the Google Perspective API, which the README says stops on December 31, 2026.
Requirements
A remote streamable HTTP endpoint; no local package or runtime. An Apify API token is required, sent as the Authorization header in the form Bearer ; the README says to get it in the Apify Console under Settings, then API & Integrations. Network access to the Apify actor endpoint is needed.
Before you install
The Authorization header carries an Apify API token, a billable credential; keep it secret and do not share it. Usage is paid: $0.50 per 1,000 text units, where a text unit is each started 1,000 characters, with lower prices on Apify paid plans. Text you submit is sent to a third-party Apify-hosted service. The first request after 60 seconds idle can take 10 to 30 seconds.

Installation

In SourceWeft

  1. Open Perspective API Alternative (Toxicity Detection) in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "perspective-api-alternative": {
      "type": "http",
      "url": "https://koyourmoon--perspective-api-alternative.apify.actor/mcp"
    }
  }
}

README

Perspective API Alternative: Toxicity Detection MCP Server

This MCP server gives AI agents a tool to examine text before they publish it. The tool returns six scores from 0 to 1 and a flagged value for each text:

  • TOXICITY, SEVERE_TOXICITY, INSULT, PROFANITY, THREAT, IDENTITY_ATTACK

The model is the open-source Detoxify model. It is not an LLM. Thus, the same text always gets the same score. English only.

The same server is also a direct replacement for the Google Perspective API. Google stops the Perspective API on December 31, 2026.

This repository holds the data that MCP directories show. The server runs on Apify: https://apify.com/koyourmoon/perspective-api-alternative

Connection

ItemValue
TransportStreamable HTTP
URLhttps://koyourmoon--perspective-api-alternative.apify.actor/mcp
AuthenticationAuthorization: Bearer <APIFY_TOKEN>

Get an Apify token in the Apify Console: Settings > API & Integrations.

Client configuration (example)

json
{  "mcpServers": {    "perspective-api-alternative": {      "type": "http",      "url": "https://koyourmoon--perspective-api-alternative.apify.actor/mcp",      "headers": { "Authorization": "Bearer <APIFY_TOKEN>" }    }  }}

Prices

  • $0.50 for 1,000 text units. A text unit is each started 1,000 characters of a text. Most comments are 1 unit.
  • Lower prices for Apify Bronze, Silver and Gold plans.
  • The first request after an idle period of 60 seconds can take 10 to 30 seconds (cold start).

Links

Source: README.md at commit b661433

Tools

0
Tool metadata has not been indexed yet.

Version history

1
  1. v1.0.9LatestOct 10, 2026