Choose your preferred integration method and make your first request.
Python
JavaScript
cURL
Install the Python SDK and tokenize your first text.
Copy
pip install blindfold-sdk
Copy
from blindfold import Blindfold# Initialize the clientclient = Blindfold(api_key="your-api-key-here")# Tokenize text with sensitive dataresponse = client.tokenize( "My email is [email protected] and phone is +1-555-1234")print(response.text)# Output: "My email is <EMAIL_ADDRESS_1> and phone is <PHONE_NUMBER_1>"print(response.mapping)# Output: {'<EMAIL_ADDRESS_1>': '[email protected]', '<PHONE_NUMBER_1>': '+1-555-1234'}
The SDK automatically handles API authentication and request formatting.
Install the JavaScript SDK and tokenize your first text.
Copy
npm install @blindfold/sdk
Copy
import { Blindfold } from '@blindfold/sdk';// Initialize the clientconst client = new Blindfold({ apiKey: 'your-api-key-here'});// Tokenize text with sensitive dataconst response = await client.tokenize( "My email is [email protected] and phone is +1-555-1234");console.log(response.text);// Output: "My email is <EMAIL_ADDRESS_1> and phone is <PHONE_NUMBER_1>"console.log(response.mapping);// Output: {'<EMAIL_ADDRESS_1>': '[email protected]', '<PHONE_NUMBER_1>': '+1-555-1234'}
The SDK works in both Node.js and browser environments.
Make a direct API call using cURL or any HTTP client.
Copy
curl -X POST https://api.blindfold.dev/api/public/v1/tokenize \ -H "X-API-Key: your-api-key-here" \ -H "Content-Type: application/json" \ -d '{ "text": "My email is [email protected] and phone is +1-555-1234" }'
text: The anonymized/protected text with sensitive data replaced
mapping: A dictionary mapping tokens back to original values (for tokenize endpoint)
entities_count: Number of sensitive entities detected
detected_entities: Detailed information about each detected entity
The mapping field is only returned by the /tokenize endpoint, as it is needed for detokenization. Other methods like /mask, /redact, and /encrypt do not return mappings.
Tokenize user input before sending to OpenAI, Anthropic, or other LLMs, then detokenize the AI response.
Copy
# Protect user inputprotected = client.tokenize(user_message)# Send to AI (safe text)ai_response = openai.chat.completions.create( messages=[{"role": "user", "content": protected.text}])# Restore original data in responsefinal_response = client.detokenize( ai_response.choices[0].message.content, protected.mapping)
Anonymizing data for analytics
Use hashing to create consistent identifiers without storing real PII.
Copy
# Hash sensitive data for analyticshashed = client.hash( "[email protected] bought product X", hash_type="sha256")# Result: "ID_a3f8b9... bought product X"
Displaying partial information
Mask sensitive data for UI display while maintaining usability.
