# TextRazor MCP for AI Agents AI Agent Connect

> TextRazor MCP lets your agent pull out structured data from messy text or websites. It identifies people, places, and companies, maps out how they're related, and categorizes content into specific topics automatically. Use it to turn raw web pages into organized datasets or to build custom terminology lists for your specific industry.

## Overview
- **Category:** knowledge-management
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_vecqfh8K0a3E1sxbBP1vh9Y5BRM4TTm2vrIW6dfx/ai-agent-connect
- **Tags:** nlp, entity-extraction, text-analysis, semantic-web, topic-tagging

## Description

TextRazor lets you connect advanced NLP to your AI agent for deep analysis of raw text or live web content. Imagine you have a mountain of news articles or a messy collection of PDFs. Usually, getting any useful data out of them requires a lot of manual copying and tagging. With this Connector, your agent does the heavy lifting. It reads through the text or visits a URL and picks out the important bits. It knows who is involved, where things are happening, and how everything connects. You can even teach it your own specific vocabulary by building custom dictionaries. Instead of hunting for keywords, you get a structured summary of what actually matters. This is a core part of the Vinkius catalog, making it easy to plug into your existing agent setup. It takes the guesswork out of content analysis by turning unstructured noise into clear, organized information you can actually use for research or reporting.

## Tools

### delete_category
Delete a custom classifier. Use this to clean up your category list when you no longer need a specific tag.

### list_dictionary_entries
List all entries in a custom dictionary. Quickly see every custom term you've added to a specific list.

### list_dictionaries
List all custom dictionaries. Use this to see an overview of all your custom-built term lists.

### analyze_text
Provide either text or url, and a comma-separated list of extractors. This is the main tool for pulling entities and topics from any source.

### create_category
Create or update a custom classifier/category. Define new topics for your agent to use when organizing content.

### create_dictionary
Create a new custom dictionary. This lets you define specific terms for your agent to watch out for.

### delete_dictionary
Delete a custom dictionary and its entries. This helps you manage your storage and remove old term lists.

### get_account
Get account details, usage, and limits. Check your remaining balance and API status to keep your agent running.

### get_dictionary
Get details of a specific custom dictionary. Use this to see what terms are currently saved in a specific list.

### list_categories
List all categories in a custom classifier. This shows you every topic your agent can currently use for tagging.

### add_dictionary_entries
Add entries to a custom dictionary. Populate your custom lists with the specific jargon of your industry.

## Prompt Examples

**Prompt:** 
```
Look at this news article and tell me which companies are mentioned and what their relationship is: [URL]
```

**Response:** 
```
I've analyzed the article. Here are the key entities and their relationships:

| Entity | Type | Relationship |
| :--- | :--- | :--- |
| **TechCorp** | Company | Primary developer of the new software |
| **Global Logistics** | Company | Primary shipping partner |
| **Sarah Jenkins** | Person | CEO of TechCorp

**Key Summary:**
* TechCorp has signed a 5-year deal with Global Logistics.
* Sarah Jenkins announced the partnership at the annual summit.
```

**Prompt:** 
```
I need to extract all the locations and people from this text: [Text]
```

**Response:** 
```
I've processed that text for you. Here are the people and locations identified:

**People:**
* John Doe (Lead Researcher)
* Maria Garcia (Field Coordinator)

**Locations:**
* **Berlin, Germany** (Main headquarters)
* **Austin, Texas** (Testing site)
* **Tokyo, Japan** (Distribution hub)
```

**Prompt:** 
```
Create a list of terms for a high-end watch brand and tell me if these appear in this product description.
```

**Response:** 
```
I've created a custom dictionary for your watch brand. 

**Terms added:**
* Chronometer
* Tourbillon
* Sapphire Crystal
* In-house Movement

**Analysis of product description:**
I found 3 matches:
1. **Tourbillon** (Mentioned in the first paragraph)
2. **Sapphire Crystal** (Mentioned in the specs section)
3. **In-house Movement** (Mentioned in the history section)
```

## Capabilities

### Extract entities from text
Your agent identifies specific people, organizations, and locations within a block of text.

### Map complex relations
It figures out how different entities in a document are connected to each other.

### Analyze web content
The agent can visit a public URL and pull out key topics and data points directly.

### Build custom dictionaries
You can create unique lists of terms for the agent to recognize in your niche.

### Classify content topics
It automatically sorts information into thousands of pre-defined categories.

## Use Cases

### Extracting entities from long reports
A researcher needs to know every company mentioned in a 50-page PDF. They ask the agent to process the text and return a structured list of organizations.

### Automated SEO category tagging
A content team wants to tag 500 old articles with new SEO categories. The agent scans the content and assigns the correct labels automatically.

### Converting reviews into structured data
A data scientist needs to turn 10,000 tweets into a structured CSV of names and locations. The agent extracts these details from every post.

### Real-time competitor monitoring
A news monitor wants to track relations between politicians and lobbyists. The agent identifies these connections as they appear in live news feeds.

## Benefits

- Automate entity extraction from thousands of documents using analyze_text to save hours of manual tagging.
- Build niche-specific knowledge bases by creating custom terms with create_dictionary for better accuracy.
- Map complex relationships between companies and people automatically to see the full picture of your industry.
- Classify massive amounts of web content into organized topics using list_categories for faster research.
- Clean up web data instantly by stripping HTML tags and focusing on the actual content during analysis.

## How It Works

The bottom line is that it turns messy human language into organized data for your agent to process.

1. Add your TextRazor API key to the Connector configuration.
2. Give your agent a text block or a website link to investigate.
3. Receive a structured list of entities, topics, and relationships.

## Frequently Asked Questions

**Can TextRazor MCP find specific people in a long article?**
Yes, it identifies names and roles within a block of text. It can pick out specific individuals even in very long documents.

**How does TextRazor MCP handle websites?**
It visits the public URL and pulls out the key data points. This means your agent can summarize and extract facts from live web pages.

**Can I teach TextRazor MCP my own industry terms?**
Yes, you can build custom dictionaries. This allows your agent to recognize and prioritize the specific jargon used in your niche.

**Is TextRazor MCP good for SEO research?**
It is excellent for SEO because it automatically tags content with relevant topics. This helps you categorize large amounts of data quickly.

**Can TextRazor MCP find relations between entities?**
Yes, it maps how different things connect. It can identify who owns what, who works for whom, and how companies are partnered.

**Does TextRazor MCP work on PDFs?**
It works on any text input. If you provide the text content from a PDF, the Connector can analyze it just like a web page or article.

**How can I analyze a web page directly without copying the text?**
Use the `analyze_text` tool and provide the target link in the `url` parameter. TextRazor will fetch the content and process it based on your requested extractors.

**Can I filter entities by specific types like DBPedia?**
Yes, the `analyze_text` tool includes an `entities_filterDbpediaTypes` parameter. You can provide a list of types to narrow down the results to exactly what you need.

**How do I manage custom terminology for my business?**
You can use `create_dictionary` to initialize a new knowledge base and then use `add_dictionary_entries` to populate it with your specific terms and IDs.