Unplugg

Overview

Unplugg offers a set of Application Programming Interfaces (APIs) engineered to assist developers and organizations in managing digital content by identifying and mitigating risks associated with text-based data. Established in 2024, Unplugg focuses on three core product areas: Content Moderation, Personally Identifiable Information (PII) Detection, and Toxicity Detection.

The Content Moderation API is designed to scan user-generated content, comments, reviews, or any textual input for violations of predefined safety guidelines or community standards. This includes detection of various categories of objectionable content such as hate speech, harassment, violence, and sexually explicit material. Its utility extends to social media platforms, online forums, e-commerce sites, and any application where user contributions require automated screening.

The PII Detection API serves to identify and extract sensitive personal data embedded within text. This capability is critical for applications handling customer data, ensuring compliance with data privacy regulations like the General Data Protection Regulation (GDPR). The API can pinpoint common PII types such as names, addresses, phone numbers, email addresses, and credit card numbers, enabling developers to anonymize, redact, or secure this information appropriately.

Complementing these, the Toxicity Detection API specifically analyzes text for expressions of aggression, insult, and other forms of toxic language. This can help platforms maintain a positive and respectful user environment by flagging or filtering out detrimental communications. Unplugg's APIs are structured to offer straightforward integration, with documentation providing clear examples for Python and Node.js SDKs, facilitating rapid deployment in various application architectures.

The service is primarily suited for use cases requiring real-time content analysis, such as live chat moderation, immediate feedback processing, or dynamic content filtering. According to the developer experience notes, the API endpoints are designed for ease of use, suggesting that developers can implement these moderation and detection capabilities with minimal complexity.

Key features

• Real-time Content Moderation: Automatically identifies and flags content violating platform guidelines across various categories including hate speech, harassment, and violence.
• PII Detection: Scans text for personally identifiable information (PII) such as names, addresses, and financial details to support data privacy and compliance efforts.
• Toxicity Detection: Analyzes textual input to detect and quantify levels of toxicity, aggression, and other forms of harmful language.
• Language Support: Designed to process content in multiple languages, enabling broad application across diverse user bases.
• Developer SDKs: Provides official client libraries for Python and Node.js to streamline integration into existing applications.
• GDPR Compliance: Adheres to General Data Protection Regulation (GDPR) standards for data processing and privacy.
• Scalable API: Supports varying levels of API call volumes, from free tier usage to high-volume enterprise plans.

Pricing

Unplugg offers a tiered pricing model, including a free tier for initial development and testing. Paid plans scale with API call volume requirements.

Pricing data as of 2026-05-28. For detailed and up-to-date information, refer to the Unplugg pricing page.

Common integrations

Unplugg's APIs are designed for integration into various application types and platforms that handle user-generated text content. Common integration points include:

• Web Applications: Backends for social media platforms, forums, and content management systems (CMS) to moderate real-time user submissions.
• Mobile Applications: Incorporating content screening directly into iOS or Android apps for user comments, chat features, and profile descriptions.
• Customer Service Platforms: Analyzing incoming customer queries or support tickets for toxicity or PII before agents respond.
• E-commerce Platforms: Moderating product reviews, seller descriptions, and customer Q&A sections.
• Internal Tools: Scanning internal communications or document repositories for sensitive information to ensure compliance.

Alternatives

Developers seeking content moderation and PII detection capabilities may consider several alternatives:

• OpenAI Moderation API: Provides a general-purpose content moderation model for identifying unsafe content categories.
• Google Cloud Content Moderation (via Perspective API): Offers advanced toxicity and attribute analysis for text, often used in large-scale content platforms.
• Amazon Rekognition (Content Moderation): Primarily focused on image and video content moderation, but AWS also offers text analysis services like Amazon Comprehend for PII detection.
• Azure AI Content Moderator: A service from Microsoft Azure that helps detect potentially offensive or unwanted text, images, and videos.
• Twilio Verify: While not a direct content moderation tool, Twilio's services can be used to verify user identities, which indirectly contributes to reducing malicious content by deterring anonymous bad actors.

Getting started

To begin using Unplugg's Content Moderation API, you typically make an HTTP POST request to the moderation endpoint with your text content. Below is a Python example demonstrating how to check a piece of text for toxicity using the Unplugg API. This example assumes you have an API key and the requests library installed.

import requests
import json

UNPLUGG_API_KEY = "YOUR_UNPLUGG_API_KEY"
UNPLUGG_API_BASE_URL = "https://api.unplugg.com/v1"

headers = {
    "Authorization": f"Bearer {UNPLUGG_API_KEY}",
    "Content-Type": "application/json"
}

def moderate_text(text_to_moderate):
    endpoint = f"{UNPLUGG_API_BASE_URL}/moderate/text"
    payload = {"text": text_to_moderate}

    try:
        response = requests.post(endpoint, headers=headers, data=json.dumps(payload))
        response.raise_for_status()  # Raise an exception for HTTP errors
        return response.json()
    except requests.exceptions.HTTPError as http_err:
        print(f"HTTP error occurred: {http_err}")
        print(f"Response: {response.text}")
    except Exception as err:
        print(f"Other error occurred: {err}")
    return None

# Example usage:
text_example_1 = "This is a perfectly normal and friendly comment."
text_example_2 = "I hate this product, it's absolutely terrible and a waste of money!"

print("Moderating text 1:")
result_1 = moderate_text(text_example_1)
if result_1:
    print(json.dumps(result_1, indent=2))

print("\nModerating text 2:")
result_2 = moderate_text(text_example_2)
if result_2:
    print(json.dumps(result_2, indent=2))

This Python script defines a function moderate_text that takes a string as input, constructs the necessary HTTP request with the API key, and sends it to Unplugg's text moderation endpoint. The response, containing moderation results, is then printed. Replace "YOUR_UNPLUGG_API_KEY" with your actual API key obtained from your Unplugg account. Refer to the Unplugg documentation for more advanced usage patterns and API endpoint details.