Name: PII Redaction API: Remove Personal Data from Text
Brand: writerslogic
SKU: api-writerslogic-pii-redaction
Availability: InStock
Author: API Market

David Condrey

# PII Redaction API

## Overview

The PII Redaction API removes personal data from text at the level you choose and then scans its own output again with the same rules. Levels are cumulative: 1 covers government identifiers (SSN, ITIN, EIN, UK NINO, Canadian SIN, driver's licence and passport), 2 adds dates of birth, 3 adds email addresses and phone numbers, and 4 adds names and street addresses. You can also pick exact categories, add or remove categories, redact your own terms, protect strings that must stay, and choose how replacements look.

It is built for developers of AI and LLM pipelines, support and CRM tools, healthcare, legal and HR software who must strip personal data before text is stored, logged or sent to a model. In our tests a patient note lost the name, date of birth, SSN, email, phone and clinic name in 2.5 seconds with the self-check passing, and a contact note in `numbered` style replaced the same person with the same `[PERSON_1]` token every time.

## Features

*   **1 endpoint**: one POST call per document.
*   **4 cumulative levels**: from government identifiers up to names and addresses.
*   **Fine control**: `categories`, `include`, `exclude`, `terms`, `keep` and `minDigitRun`.
*   **Replacement styles**: `label`, `generic`, `numbered` and `last4`.
*   **Self-check**: `verification` reports residual detections and leaked values after redaction.
*   **Explained results**: every redaction lists its category, position, confidence and rule.
*   **Pay per call**: billed per document on the PAYG plan.

## Use Cases

*   **LLM pipelines**: remove personal data before sending text to a model.
*   **Support tickets**: redact customer details before storing or sharing tickets.
*   **Healthcare and legal**: prepare documents for review or research.
*   **Logs**: scrub personal data from application logs.
*   **Data sharing**: anonymise exports before they leave your system.

## API Documentation

### Base URL

All requests use the following base URL.

```
https://prod.api.market/api/v1/writerslogic/pii-redaction
```

### Authentication

All requests are authenticated with the `x-api-market-key` header.

```
x-api-market-key: YOUR_API_KEY
```

Replace `YOUR_API_KEY` with your actual API key.

### Endpoints

#### 1. Redact text

Redacts personal data from `text` and returns the redacted text, a summary, every redaction with its position and rule, and the self-check result.

**Method:** `POST`

**Endpoint:** `/v1/redact-text`

**Parameters:**

| Name | In | Description | Type | Required | Default | Max Value |
|---|---|---|---|---|---|---|
| `text` | body | Text to redact: plain text, Markdown, CSV, JSON or HTML source | string | Yes | - | - |
| `level` | body | Cumulative preset from 1 to 4 | integer | No | `1` | 4 |
| `categories` | body | Exact set of categories, replacing the level | array | No | - | - |
| `include` | body | Categories to add to the level, for example `["organization"]` | array | No | - | - |
| `exclude` | body | Categories to remove from the level | array | No | - | - |
| `terms` | body | Words or phrases to redact wherever they appear | array | No | - | - |
| `keep` | body | Strings that must never be redacted | array | No | - | - |
| `minDigitRun` | body | Also redact every run of at least this many digits | integer | No | - | - |
| `style` | body | `label`, `generic`, `numbered` or `last4` | string | No | `label` | - |
| `useModel` | body | Add a language model pass for names, organisations and addresses | boolean | No | `false` | - |

**Example Request:**

```bash
curl -X POST 'https://prod.api.market/api/v1/writerslogic/pii-redaction/v1/redact-text' \
  -H 'x-api-market-key: YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"text": "Patient: Maria Gonzalez, DOB 04/12/1980, SSN 123-45-6789. Reach her at maria.gonzalez@example.org or (555) 010-0199. Maria Gonzalez was seen at Acme Clinic, Inc.", "level": 4, "include": ["organization"], "style": "label"}'
```

**Example Response (200 OK):**

```json
{
  "data": {
    "redactedText": "Patient: [REDACTED:PERSON], DOB [REDACTED:DOB], SSN [REDACTED:SSN]. Reach her at [REDACTED:EMAIL] or [REDACTED:PHONE]. [REDACTED:PERSON] was seen at [REDACTED:ORG]",
    "status": "complete",
    "summary": {
      "total": 7,
      "byCategory": {
        "person_name": 2,
        "date_of_birth": 1,
        "ssn": 1,
        "email": 1,
        "phone": 1,
        "organization": 1
      },
      "charactersRemoved": 106
    },
    "redactions": [
      {
        "category": "person_name",
        "start": 9,
        "end": 23,
        "length": 14,
        "replacement": "[REDACTED:PERSON]",
        "confidence": "context",
        "rule": "name-after-patient"
      },
      {
        "category": "date_of_birth",
        "start": 29,
        "end": 39,
        "length": 10,
        "replacement": "[REDACTED:DOB]",
        "confidence": "context",
        "rule": "dob-label"
      },
      {
        "category": "ssn",
        "start": 45,
        "end": 56,
        "length": 11,
        "replacement": "[REDACTED:SSN]",
        "confidence": "validated",
        "rule": "ssn-formatted"
      }
    ],
    "verification": {
      "passed": true,
      "residualDetections": 0,
      "leakedValues": 0,
      "distinctValues": 6,
      "outputSha256": "12cba160d18d9128c795727caa3751a2729b254e992349f815e8fc238f5abe34"
    },
    "config": {
      "level": 4,
      "categories": [
        "address",
        "date_of_birth",
        "drivers_license",
        "ein",
        "email",
        "itin",
        "nino",
        "organization",
        "passport",
        "person_name",
        "phone",
        "sin",
        "ssn"
      ],
      "minDigitRun": null,
      "terms": 0,
      "keep": 0,
      "style": "label",
      "useModel": false
    },
    "coverage": [
      {
        "category": "address",
        "count": 0,
        "method": "pattern",
        "limits": "US street, PO box, city-state-ZIP and UK postcode layouts. Other countries' addresses are not covered; use useModel for those."
      },
      {
        "category": "date_of_birth",
        "count": 1,
        "method": "label-keyed",
        "limits": "Dates within a few characters after DOB, date of birth, born. An unlabeled birth date is only caught with the date category."
      }
    ],
    "limits": [
      "Detection finds what its rules and your terms describe; it cannot prove a document contains no other personal data. Review the output before publishing.",
      "Plain text only. For PDF, DOCX or images, extract the text, redact it, and rebuild the file; a visual overlay does not remove text from the file.",
      "No model was used and the text left this service nowhere."
    ],
    "deterministic": true
  },
  "_provenance": {
    "version": "writerslogic-provenance/1",
    "issuer": "did:web:agents.writerslogic.com",
    "verify": {
      "endpoint": "POST /api/v1/verify-provenance",
      "did": "did:web:agents.writerslogic.com"
    },
    "manifest": {
      "format": "application/json",
      "instance_id": "xmp:iid:b35248d3-93c1-413e-bacf-5f0cbc9ccdf8",
      "claim_generator_info": [
        {
          "name": "WritersLogic x402 Agent Hub",
          "version": "2.0.0"
        }
      ],
      "title": "Selective PII redaction with a self-check",
      "assertions": [
        {
          "label": "c2pa.actions.v2",
          "data": {
            "actions": [
              {
                "action": "c2pa.created",
                "when": "2026-11-11T20:27:59.623Z",
                "softwareAgent": {
                  "name": "writerslogic-agent-hub/redact-text",
                  "version": "2.0.0"
                },
                "digitalSourceType": "http://cv.iptc.org/newscodes/digitalsourcetype/algorithmicMedia",
                "parameters": {
                  "endpoint": "POST /api/v1/redact-text"
                }
              }
            ]
          }
        },
        {
          "label": "c2pa.hash.data",
          "data": {
            "alg": "sha256",
            "hash": "7581017ccdbe623929ede242a7835ca05dcb1f86664e7a705cab10704263cd88",
            "name": "jcs-canonical-json(data)"
          }
        }
      ]
    },
    "credential": {
      "@context": [
        "https://www.w3.org/ns/credentials/v2"
      ],
      "id": "urn:uuid:598bd868-e106-4658-80d5-c55fb26a875b",
      "type": [
        "VerifiableCredential",
        "AgentOutputCredential"
      ],
      "issuer": "did:web:agents.writerslogic.com",
      "validFrom": "2026-11-11T20:27:59.623Z",
      "credentialSubject": {
        "id": "urn:sha256:7581017ccdbe623929ede242a7835ca05dcb1f86664e7a705cab10704263cd88",
        "agent": "redact-text",
        "endpoint": "POST /api/v1/redact-text",
        "outputDigest": {
          "alg": "sha256",
          "value": "7581017ccdbe623929ede242a7835ca05dcb1f86664e7a705cab10704263cd88",
          "canonicalization": "JCS"
        },
        "manifestDigest": {
          "alg": "sha256",
          "value": "740ae4f5b83f935c498db253b4255a172d864bc08a9bec3bebc4a7d0ad5fcb05",
          "canonicalization": "JCS"
        },
        "ingredients": [],
        "payment": {
          "scheme": "api-key",
          "network": "eip155:8453",
          "payer": "partner:api-market"
        }
      },
      "proof": {
        "type": "DataIntegrityProof",
        "cryptosuite": "eddsa-jcs-2022",
        "created": "2026-11-11T20:27:59.623Z",
        "verificationMethod": "did:web:agents.writerslogic.com#key-1",
        "proofPurpose": "assertionMethod",
        "proofValue": "z3Ho2QiAWK8sd9jLtVVdcGrKPLo5guFWrG65VTQZyNy9XWgVQC2uDABEhcj3QY1mjMUGMbgk2SAiNmnNrd6bDWVFX"
      }
    },
    "ethSignature": {
      "type": "EIP-191",
      "signer": "did:pkh:eip155:8453:0x3DBd42560cA7a7E448A2c7DfC9E785526189411f",
      "address": "0x3DBd42560cA7a7E448A2c7DfC9E785526189411f",
      "message": "writerslogic-provenance:acfdcab4392273fee07f27435b0f3c4b9c2ec3debeb26a09ad70fce0490697f1",
      "signature": "0x5885ef72860929965b0c7c1ce4edd31d36dfe97abd76c927e16fd54d2bd9445566c630f239f02e7f822bb37ea395a01475cca6af6675afd2e9a85578d1a206a31c"
    }
  }
}
```

The `redactions` and `coverage` lists above are shortened. With `style: numbered` the same value always gets the same token:

```json
{
  "data": {
    "redactedText": "Contact [PERSON_1] at [EMAIL_1], card 4111 1111 1111 1111, IBAN GB82 WEST 1234 5698 7654 32. Ship to [ADDRESS_1], London [ADDRESS_2].",
    "status": "complete",
    "summary": {
      "total": 4,
      "byCategory": {
        "person_name": 1,
        "email": 1,
        "address": 2
      },
      "charactersRemoved": 55
    }
  },
  "_provenance": {
    "version": "writerslogic-provenance/1",
    "issuer": "did:web:agents.writerslogic.com",
    "verify": {
      "endpoint": "POST /api/v1/verify-provenance",
      "did": "did:web:agents.writerslogic.com"
    },
    "manifest": {
      "format": "application/json",
      "instance_id": "xmp:iid:66d42e30-b69c-4da7-9542-c658d2be5543",
      "claim_generator_info": [
        {
          "name": "WritersLogic x402 Agent Hub",
          "version": "2.0.0"
        }
      ],
      "title": "Selective PII redaction with a self-check",
      "assertions": [
        {
          "label": "c2pa.actions.v2",
          "data": {
            "actions": [
              {
                "action": "c2pa.created",
                "when": "2026-11-11T20:28:14.380Z",
                "softwareAgent": {
                  "name": "writerslogic-agent-hub/redact-text",
                  "version": "2.0.0"
                },
                "digitalSourceType": "http://cv.iptc.org/newscodes/digitalsourcetype/algorithmicMedia",
                "parameters": {
                  "endpoint": "POST /api/v1/redact-text"
                }
              }
            ]
          }
        },
        {
          "label": "c2pa.hash.data",
          "data": {
            "alg": "sha256",
            "hash": "bab3cab5565cb8614b2238afb2904f9e2added40238284a3f79b6f1eb977ad71",
            "name": "jcs-canonical-json(data)"
          }
        }
      ]
    },
    "credential": {
      "@context": [
        "https://www.w3.org/ns/credentials/v2"
      ],
      "id": "urn:uuid:6fde4f47-75fa-48ba-b076-8eca2e0248cb",
      "type": [
        "VerifiableCredential",
        "AgentOutputCredential"
      ],
      "issuer": "did:web:agents.writerslogic.com",
      "validFrom": "2026-11-11T20:28:14.380Z",
      "credentialSubject": {
        "id": "urn:sha256:bab3cab5565cb8614b2238afb2904f9e2added40238284a3f79b6f1eb977ad71",
        "agent": "redact-text",
        "endpoint": "POST /api/v1/redact-text",
        "outputDigest": {
          "alg": "sha256",
          "value": "bab3cab5565cb8614b2238afb2904f9e2added40238284a3f79b6f1eb977ad71",
          "canonicalization": "JCS"
        },
        "manifestDigest": {
          "alg": "sha256",
          "value": "e4e338deec219346ab086a099b84e9934e1ad33616e1ba42638f3c41e3645043",
          "canonicalization": "JCS"
        },
        "ingredients": [],
        "payment": {
          "scheme": "api-key",
          "network": "eip155:8453",
          "payer": "partner:api-market"
        }
      },
      "proof": {
        "type": "DataIntegrityProof",
        "cryptosuite": "eddsa-jcs-2022",
        "created": "2026-11-11T20:28:14.380Z",
        "verificationMethod": "did:web:agents.writerslogic.com#key-1",
        "proofPurpose": "assertionMethod",
        "proofValue": "z5Q6chJaZ2S9RcirPFXuzEaYmdYi6awuZRZ1dysBWpbQsPNs3x3yTNGy4o3idkHPV91mp5BxGSncmypP9aHzZKjrJ"
      }
    },
    "ethSignature": {
      "type": "EIP-191",
      "signer": "did:pkh:eip155:8453:0x3DBd42560cA7a7E448A2c7DfC9E785526189411f",
      "address": "0x3DBd42560cA7a7E448A2c7DfC9E785526189411f",
      "message": "writerslogic-provenance:a9b55f5b5c2ff18cca6affcf0ba22bb83f05dadafb9df5a8ac3ad5de2ab69ab6",
      "signature": "0x3eb3a2789c90e4bee4c29fa407a844520899bea5f18cac218f3321f1f52674fc672d8d7e289ab286879c193697f044c2ccaeaad7754b6c7bffba5f9e6bf3ebd41b"
    }
  }
}
```

Card numbers and IBANs are not part of levels 1 to 4. Use `minDigitRun` or `terms` to remove them.

## Status Codes

| Code | Status | Description |
|---|---|---|
| 200 | OK | The request succeeded and the JSON result is in the response body. |
| 400 | Bad Request | The body is invalid, for example `{"error": "$.level: above 4", "code": "invalid_request"}`. |
| 401 | Unauthorized | The `x-api-market-key` header is missing or the key is not valid. |
| 402 |  | Your API.market balance cannot cover the call. |
| 403 | Forbidden | Your key does not have an active subscription that includes this endpoint. |
| 404 | Not Found | The path does not match an endpoint of this API. Check the Base URL and path. |
| 429 | Too Many Requests | You went over 10 requests per second. |
| 500 | Internal Server Error | The request failed on the server. Retry once, then contact us if it continues. |
| 503 | Service Unavailable | The service is temporarily unavailable. Retry after a short wait. |

## Rate Limits

The PAYG plan allows up to 10 requests per second and is billed per call. When you go over a limit, the API returns `429 Too Many Requests`. Throttle your requests, back off before retrying, and honour the `Retry-After` header when it is present.

## FAQ

*   **Which level should I use?**
    Level 4 covers most documents. Use a lower level when you only need to remove identifiers, and `include` to add organisations.

*   **Does it remove card numbers?**
    Not with the levels alone. Add `minDigitRun`, for example 12, to remove long digit runs such as card and account numbers.

*   **Does the self-check prove all personal data is gone?**
    No. It rescans the output with the same rules. Data the rules do not cover can remain, so review sensitive documents.

*   **Is my text stored?**
    Text is processed for the request. Turn on `useModel` only if you accept a model pass over the text.

## Contact Us

If you have any questions or need support, reach out through the chat embedded on this page or email us at hello@magicapi.com.


Redact names, emails, phone numbers, SSNs, dates of birth and addresses from text at the level you choose, with a second scan of the output.