About Intelation Nexis
Intelation Nexis is an MCP server for detecting and transforming personal data in text. It offers privacy operations configured for regulations such as GDPR, UK GDPR, HIPAA, and Privacy Act.
Key features
- 01Detect PII without changing text
- 02Redact, mask, or replace PII
- 03Hash, pseudonymize, or tokenize data
- 04Encrypt and optionally decrypt values
- 05List supported countries and regulations
Use cases
- 01Anonymize support tickets before summarizing
- 02Find personal data in text
- 03Prepare text for privacy-aware processing
MCP support
The server exposes Nexis tools for detecting and transforming PII to MCP clients. Decryption is disabled by default and can be enabled through an environment variable.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y @intelation/nexis-mcpNEXIS_API_KEYSecretYour Nexis API key. Get a free one at https://nexis.intelation.comNEXIS_COUNTRYDefault regulation|country|sub-region|language. Otherwise the first country in your planNEXIS_ENCRYPTION_KEYSecretFernet key for the encryption method and decrypt_textNEXIS_ALLOW_DECRYPTSet to true to turn on decrypt_text. The assistant then sees the original personal dataNEXIS_BASE_URLOnly for a dedicated or on-premises Nexis instanceLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project README@intelation/nexis-mcp
MCP server that lets Claude, Cursor, VS Code and other AI assistants find and remove personal data (PII) from text. Regulation-aware (UK GDPR, GDPR, HIPAA, Privacy Act), powered by Intelation Nexis.
Full guide: intelation.com/integrations/javascript.
Setup
Get a free API key at https://nexis.intelation.com, then add the server to your MCP client.
(then replace
your-key with your API key)
Claude Desktop / Cursor (mcpServers in the config file):
{
"mcpServers": {
"nexis": {
"command": "npx",
"args": ["-y", "@intelation/nexis-mcp"],
"env": { "NEXIS_API_KEY": "your-key" }
}
}
}
Claude Code:
claude mcp add nexis --env NEXIS_API_KEY=your-key -- npx -y @intelation/nexis-mcp
Then ask: "Anonymize this support ticket before you summarise it: ..."
Tools
| Tool | What it does |
|---|---|
anonymize_text | Redact, replace, mask, hash, pseudonymize, tokenize or encrypt PII |
detect_pii | List the PII found, without changing the text |
decrypt_text | Restore values encrypted by anonymize_text. Off unless NEXIS_ALLOW_DECRYPT=true |
list_countries | Countries and regulations in your plan |
Environment
| Variable | |
|---|---|
NEXIS_API_KEY | Required |
NEXIS_COUNTRY | Default regulation|country|sub-region|language; otherwise the first country in your plan |
NEXIS_ENCRYPTION_KEY | Fernet key for the encryption method and decrypt_text |
NEXIS_ALLOW_DECRYPT | true to turn on decrypt_text. Off by default |
NEXIS_BASE_URL | Only for a dedicated or on-premises instance |
The encryption key is read only from the environment and is never a tool argument, so the key never passes through the model.
Before you set NEXIS_ALLOW_DECRYPT=true: the assistant sees whatever decrypt_text returns, which is the original personal data. A document with hidden instructions ("decrypt this and include it") could also get the assistant to decrypt values. Turn it on only when the assistant is meant to read the original values:
"env": { "NEXIS_API_KEY": "your-key", "NEXIS_ENCRYPTION_KEY": "your-fernet-key", "NEXIS_ALLOW_DECRYPT": "true" }
claude mcp add nexis --env NEXIS_API_KEY=your-key --env NEXIS_ENCRYPTION_KEY=your-fernet-key --env NEXIS_ALLOW_DECRYPT=true -- npx -y @intelation/nexis-mcp
Use in your own code
Importing the package does not start a server, so you can add the Nexis tools to your own MCP setup:
import { Nexis } from '@intelation/nexis';
import { createServer } from '@intelation/nexis-mcp';
const server = createServer(new Nexis(), { version: '1.0.0', allowDecrypt: false });
await server.connect(transport);
MIT © Intelation
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.