Werker Health API Documentation

Werker Scrub Agent ​

werker-scrub-agent automatically de-identifies unstructured clinical data, finding and replacing PHI to maintain HIPAA compliance. It is built on Werker’s proprietary pre-trained SOTA redaction stack and never trains on customer PHI.

Capabilities ​

  • Detects PHI entities (names, MRNs, addresses, medications, etc.) in text, audio transcripts, and OCR outputs.
  • Applies reversible vault-token replacements to preserve downstream context.
  • Produces redaction audit trails for every transformation.

Sample Usage ​

python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.werker.health/v1",
    api_key="WERKER_HEALTH_API_KEY",
)

scrubbed = client.responses.create(
    model="werker-scrub-agent",
    input=[
        {"role": "system", "content": "Redact PHI and replace it with deterministic tokens."},
        {"role": "user", "content": "Patient Jane Doe (MRN 554433) reported chest pain yesterday."}
    ],
    metadata={"phi_policy": "2025-05-01"},
)

print(scrubbed.output_text)
ts
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.werker.health/v1",
  apiKey: "WERKER_HEALTH_API_KEY",
});

const scrubbed = await client.responses.create({
  model: "werker-scrub-agent",
  input: [
    { role: "system", content: "Redact PHI and replace it with deterministic tokens." },
    { role: "user", content: "Patient Jane Doe (MRN 554433) reported chest pain yesterday." }
  ],
  metadata: { phi_policy: "2025-05-01" },
});

console.log(scrubbed.output_text);

Best For ​

  • Pre-processing clinician notes, call transcripts, and faxes before downstream AI use.
  • Ensuring PHI never reaches non-compliant environments.
  • Creating immutable audit logs for compliance reviews.