Getting Started

Learn how to get started with Quba, from signing up to exploring the playground to creating your API key and integrating with our SDK.

Create an API key and get started

In the sign-up page, enter your email and you will receive a code. Enter the code received, click verification, and you are in!

Preview text scan and anonymization in Quba

Once you are in, you land directly in the playground, where you can type your data and choose whether to scan it or anonymize it with different rules and entities.

Create your API key

Go to API Keys and click Create API key
Set a name and expiration date for your API key
Copy your API key

Now you are ready to integrate with Quba.

SDK Installation

npm install @quba/sensitive-data-protection --save

pip install quba-sensitive-data-protection
# or with uv:
uv add quba-sensitive-data-protection

This example walks through a complete flow: scanning a text for sensitive entities, or anonymizing it using per-entity rules.

The Input

const text = `
  Patient John Smith (SSN: 123-45-6789) was admitted to Dubai General Hospital
  on 2024-01-15. Contact: john.smith@email.com, +971-50-123-4567.
  Assigned physician: Dr. Sarah Al Mansoori (ID: EMP-00421).
`

text = """
  Patient John Smith (SSN: 123-45-6789) was admitted to Dubai General Hospital
  on 2024-01-15. Contact: john.smith@email.com, +971-50-123-4567.
  Assigned physician: Dr. Sarah Al Mansoori (ID: EMP-00421).
"""

Scan

Scan your text and discrover what sensitive entities it contains.

import {
  Configuration,
  SensitiveDataProtectionApi,
} from "@quba/sensitive-data-protection"

// Pass your API key on every request via the x-api-key header.
const api = new SensitiveDataProtectionApi(
  new Configuration({ headers: { "x-api-key": "quba_..." } }),
)

const scan = await api.scanText({
  text, // text to scan
  language: "en",
  entities: ["person", "email", "phone", "location", "id"],
  confidence_threshold: 0.5,
})

console.log(scan.results)
// [
//   { start: 11, end: 21, score: 0.95, entity_type: "person"   },  // John Smith
//   { start: 24, end: 35, score: 0.99, entity_type: "id"       },  // 123-45-6789
//   { start: 53, end: 74, score: 0.91, entity_type: "location" },  // Dubai General Hospital
//   { start: 93, end: 117, score: 0.98, entity_type: "email"   },  // john.smith@email.com
//   { start: 119, end: 134, score: 0.97, entity_type: "phone"  },  // +971-50-123-4567
//   { start: 158, end: 178, score: 0.93, entity_type: "person" },  // Dr. Sarah Al Mansoori
//   { start: 184, end: 193, score: 0.99, entity_type: "id"     },  // EMP-00421
// ]

from quba_sdp import Client, models
from quba_sdp.api import scan_text

# base_url defaults to the production API; pass your key via headers.
client = Client(headers={"x-api-key": "quba_..."})

scan = scan_text.sync(
    client=client,
    body=models.ScanRequestBody(
        text=text,  # text to scan
        language="en",
        entities=["person", "email", "phone", "location", "id"],
        confidence_threshold=0.5,
    ),
)

print(scan.results)
# [
#   ScanResult(start=11,  end=21,  score=0.95, entity_type="person"),    # John Smith
#   ScanResult(start=24,  end=35,  score=0.99, entity_type="id"),        # 123-45-6789
#   ScanResult(start=53,  end=74,  score=0.91, entity_type="location"),  # Dubai General Hospital
#   ScanResult(start=93,  end=117, score=0.98, entity_type="email"),     # john.smith@email.com
#   ScanResult(start=119, end=134, score=0.97, entity_type="phone"),     # +971-50-123-4567
#   ScanResult(start=158, end=178, score=0.93, entity_type="person"),    # Dr. Sarah Al Mansoori
#   ScanResult(start=184, end=193, score=0.99, entity_type="id"),        # EMP-00421
# ]

Anonymize

Apply a rule per entity type. Different fields get different treatment depending on how the output will be used.

const response = await api.anonymizeText({
  text, // text to scan and anonymize
  rules: [
    {
      type: "replace",
      entities: [{ type: "model", value: "person" }],
      replacement: "[PERSON]",
    },
    {
      type: "mask",
      entities: [{ type: "regex", value: "\\d{3}-\\d{2}-\\d{4}" }],
      masking_char: "*",
      chars_to_mask: 7,
      from_end: true,
    },
    {
      type: "redact",
      entities: [{ type: "model", value: "location" }],
    },
    {
      type: "sha256",
      entities: [{ type: "model", value: "email" }],
    },
    {
      type: "mask",
      entities: [{ type: "model", value: "phone" }],
      chars_to_mask: 8,
      from_end: true,
    },
    {
      type: "replace",
      entities: [{ type: "regex", value: "EMP-\\d+" }],
      replacement: "[EMP-ID]",
    },
  ],
})

from quba_sdp.api import anonymize_text

response = anonymize_text.sync(
    client=client,
    body=models.AnonymizeRequestBody(
        text=text,  # text to scan and anonymize
        rules=[
            models.ReplaceRule(
                entities=[models.ModelEntity(value="person")],
                replacement="[PERSON]",
            ),
            models.MaskRule(
                entities=[models.RegexEntity(value=r"\d{3}-\d{2}-\d{4}")],
                masking_char="*",
                chars_to_mask=7,
                from_end=True,
            ),
            models.RedactRule(entities=[models.ModelEntity(value="location")]),
            models.SHA256Rule(entities=[models.ModelEntity(value="email")]),
            models.MaskRule(
                entities=[models.ModelEntity(value="phone")],
                chars_to_mask=8,
                from_end=True,
            ),
            models.ReplaceRule(
                entities=[models.RegexEntity(value=r"EMP-\d+")],
                replacement="[EMP-ID]",
            ),
        ],
    ),
)

response.text is the anonymized string. response.results is the full audit trail — one record per transformation, with positions in both the original and anonymized text.

console.log(response.text)
// Patient [PERSON] (SSN: ***-**-6789) was admitted to [REDACTED]
// on 2024-01-15. Contact: a1b2c3d4e5f6..., +971-50-***-****
// Assigned physician: [PERSON] (ID: [EMP-ID]).

console.log(response.results)
// [
//   { type: "model", rule: "replace", value: "PERSON",
//     input: { start: 11, end: 21, value: "John Smith" },
//     output: { start: 11, end: 19, value: "[PERSON]" } },
//
//   { type: "regex", rule: "mask", value: "\\d{3}-\\d{2}-\\d{4}",
//     input: { start: 24, end: 35, value: "123-45-6789" },
//     output: { start: 22, end: 33, value: "***-**-6789" } },
//   ...
// ]

print(response.text)
# Patient [PERSON] (SSN: ***-**-6789) was admitted to [REDACTED]
# on 2024-01-15. Contact: a1b2c3d4e5f6..., +971-50-***-****
# Assigned physician: [PERSON] (ID: [EMP-ID]).

print(response.results)
# [
#   ModelResult(type="model", rule="replace", value="PERSON",
#     input=TextRange(start=11, end=21, value="John Smith"),
#     output=TextRange(start=11, end=19, value="[PERSON]")),
#
#   RegexResult(type="regex", rule="mask", value="\\d{3}-\\d{2}-\\d{4}",
#     input=TextRange(start=24, end=35, value="123-45-6789"),
#     output=TextRange(start=22, end=33, value="***-**-6789")),
#   ...
# ]

Use input offsets to map back to the original text. Use output offsets to highlight protected spans in the anonymized result.

Error Handling

try {
  const response = await api.scanText({ text: "" })
} catch (error) {
  if (error.status === 422) {
    console.error("Validation error:", error.detail)
  }
}

from quba_sdp.models import HTTPValidationError

# `sync` returns the parsed body, a documented error model, or None.
result = scan_text.sync(client=client, body=models.ScanRequestBody(text=""))
if isinstance(result, HTTPValidationError):
    print("Validation error:", result.detail)

# Or use `sync_detailed` when you need the raw status code:
resp = scan_text.sync_detailed(client=client, body=models.ScanRequestBody(text=""))
print(resp.status_code)  # 422

Getting Started

On this page