• PRECISION
  • CONTROL
  • CONTEXT
  • HUMAN

How to build custom industry dictionaries with your AI

Connect TextRazor to your AI client to give it a specialized vocabulary. Once connected, your agent manages its own set of industry terms.

Ask AI about this page

Short answer

How can I teach my AI niche terminology?

You can create and manage specialized dictionaries that your agent uses to identify jargon. This allows your AI to interpret technical terms that aren't in its standard training data. You'll get precise entity recognition for your specific field.

The outcomes

What a custom vocabulary provides

Giving your agent specialized knowledge changes how it reads your data.

PRECISION

Accurate jargon detection

The AI uses your custom terms to correctly identify niche entities in text. This prevents the agent from misinterpreting technical acronyms.

CONTROL

Managed term lists

You keep full control over what the AI knows. You can add, remove, or check entries to keep the vocabulary current.

CONTEXT

Better entity recognition

Your agent understands the relationship between specific terms in your industry. This leads to smarter analysis of your documents.

HUMAN

Manual oversight

You decide which terms are important enough to include. The AI only knows what you explicitly add to the dictionaries.

The workflow

What your AI does when the terminology arrives.

Your AI handles the heavy lifting of organizing and storing your specialized vocabulary.

  1. Initialize the dictionary

    Your agent sets up a new, empty container for your specific industry terms.

    create_dictionary

  2. Populate the terms

    The AI adds specific words, phrases, or acronyms to the newly created dictionary.

    add_dictionary_entries

  3. Verify the contents

    The agent checks the current list of terms to ensure everything was added correctly.

    list_dictionary_entries

  4. Retrieve specific data

    The AI pulls the full details of a specific dictionary to use during text analysis.

    get_dictionary

Try it

Copy these to start.

Use these prompts to kick off the dictionary building process.

Starting points

Swap out the bracketed text for your actual industry or specific terms.

TextRazor Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_vecqfh8K0a3E1sxbBP1vh9Y5BRM4TTm2vrIW6dfx/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — TextRazor capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "textrazor-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_vecqfh8K0a3E1sxbBP1vh9Y5BRM4TTm2vrIW6dfx/mcp"
    }
  }
}
Copy into chat04

  • Create a new dictionary called [Medical Terminology] for my research project.

  • Add [CRISPR] and [nucleotide] to my [Medical Terminology] dictionary.

  • List all the entries currently in my [Medical Terminology] dictionary.

  • Get the details for the [Legal Jargon] dictionary so I can review it.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect TextRazor once, then ask.

Just click to connect. Your credentials stay encrypted, and you can immediately start talking to your agent about your custom terms.

Connect TextRazor to your AI

FAQ

How the dictionary tools behave

  • 01

    Can the AI delete entries automatically?

    No. The AI can only add or list entries based on the available tools. You must manage deletions manually.

  • 02

    Does the AI learn from every word it sees?

    No. The AI only recognizes terms that you have explicitly added to a dictionary via the connector.

  • 03

    Can I use one dictionary for multiple AI agents?

    Yes, once the dictionary is created, any agent connected to this MCP can access it.

  • 04

    Can the AI create a dictionary from a PDF?

    No. You need to tell the AI which specific terms to add to the dictionary after it reads the document.

  • 05

    Will this change the underlying TextRazor model?

    No. This only creates a custom layer of terminology for your agent to use during its specific tasks.

  • More questions about TextRazor? See everything the TextRazor Connector can do

Connect TextRazor to Claude, Cursor, ChatGPT & more