# Unprofane [Health: Active]

**Category:** 🔒 Security  
**Repository:** https://unprofane.com/docs/mcp/  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/unprofane

## Description
Should I allow this text? Profanity and toxicity checks that are hard to fool, with the evidence.

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: medium). Install path inferred — verify against the README before running it. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "unprofane": {
    "url": "https://unprofane.com/docs/mcp/"
  }
}
```

## Documentation

## What the Unprofane MCP server does

Unprofane MCP server provides profanity and toxicity checks for text. Its stated purpose is to help answer whether a piece of text should be allowed, rather than returning only an unexplained pass or fail result. The service also provides evidence with its checks, so an application or agent can inspect why the text was flagged or accepted.

The available listing places Unprofane in the 🔒 Security category. That makes it relevant to systems where user-generated or agent-generated text must be reviewed before it is displayed, stored, or passed to another workflow. The provided material does not define a particular programming language, moderation policy, threshold, or supported text format.

## Tools and capabilities

The documented capabilities are limited to two related checks:

- Detect profanity in submitted text.
- Detect toxicity in submitted text.
- Return evidence associated with the check.

The description characterizes these checks as difficult to fool, but it does not provide benchmark results, detection rules, language coverage, false-positive rates, or examples of returned evidence. It also does not list formal MCP tool names or describe whether the checks are exposed as separate tools or as options on one tool. Treat the capability list as functional guidance rather than an API specification.

## Limitations and notes

The available source material is minimal. It does not document installation steps, environment variables, authentication, request and response schemas, transport selection, rate limits, supported clients, or licensing. As a result, the Unprofane MCP server cannot be matched to a particular editor or agent runtime from the supplied information alone.

The listing also does not state whether text is retained, how evidence is generated, whether checks run locally or remotely, or which languages are supported. Those details matter when evaluating the server for production moderation, privacy-sensitive content, or multilingual applications and should be confirmed in the linked documentation before deployment.

Use the Unprofane MCP server when the required decision is specifically about profanity or toxicity and an evidence-bearing result is useful. Do not assume it covers broader content safety categories, policy enforcement, redaction, or user reporting; none of those capabilities are described in the supplied material.

_Full upstream README: https://allmcps.com/mcp/unprofane/readme_