Skip to content

Python SDK¶

The xybern-redact package is a thin Python client for the Xybern Redact API. Install it with pip and call the same anonymization, vault, and proxy endpoints without writing any HTTP code yourself.

pip install xybern-redact

Requirements¶

Python 3.9+
A Xybern workspace API key (xr_live_...) from the API Keys tab in your dashboard

Quick start¶

from xybern_redact import RedactClient

client = RedactClient(api_key="xr_live_YOUR_KEY")

result = client.anonymize("Michael Chen signed the contract on 12 March 2024.")
print(result.text)
# "Morgan Ross signed the contract on 11 April 2024."

print(result.entity_count)   # 2
print(result.entities)       # {"PERSON": 1, "DATE": 1}

Anonymize then call any LLM yourself¶

Strip PII before sending to an LLM, then restore real values in the response:

from xybern_redact import RedactClient
from anthropic import Anthropic

redact = RedactClient(api_key="xr_live_YOUR_KEY")
llm    = Anthropic(api_key="sk-ant-...")

anon = redact.anonymize("Summarise the contract signed by Michael Chen at Goldman Sachs.")

response = llm.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=512,
    messages=[{"role": "user", "content": anon.text}],
)

original = redact.deanonymize(response.content[0].text, anon.entity_map)
print(original)
# Michael Chen and Goldman Sachs are restored.

Use the built-in proxy¶

The proxy handles anonymization and de-anonymization automatically - you send the original text and get the original names back in the response:

response = client.chat(
    model="claude-sonnet-4-6",
    messages=[{"role": "user", "content": "Summarise the contract signed by Michael Chen."}],
)
print(response.content)
# Response already has real names restored.

Anonymize a file¶

result = client.anonymize_file("contract.pdf")

with open("contract_clean.pdf", "wb") as f:
    f.write(result.content)

print(f"{result.entity_count} entities anonymized")

Multi-turn conversations¶

Pass the same thread_id across calls to keep pseudonyms consistent throughout a conversation:

thread = "session_abc123"

r1 = client.anonymize("Michael Chen signed.", thread_id=thread)
r2 = client.anonymize("Chen reviewed the annexes.", thread_id=thread)

# Both map "Chen" to the same pseudonym across the thread.

Error handling¶

from xybern_redact import RedactClient, AuthenticationError, APIError

try:
    client = RedactClient(api_key="xr_live_YOUR_KEY")
    result = client.anonymize(text)
except AuthenticationError:
    print("Invalid or missing API key.")
except APIError as e:
    print(f"Server error {e.status_code}: {e}")

Context manager¶

with RedactClient(api_key="xr_live_YOUR_KEY") as client:
    result = client.anonymize("...")

Return types¶

`AnonymizeResult`¶

Field	Type	Description
`text`	`str`	Anonymized text
`entity_map`	`dict[str, str]`	Pseudonym → real value mapping for de-anonymization
`entity_count`	`int`	Total entities replaced
`entities`	`dict[str, int]`	Count per entity type
`leakage_detected`	`bool`	Whether PII leakage was found

`ChatResult`¶

Field	Type	Description
`content`	`str`	Response text with real values restored
`model`	`str`	Model used
`usage`	`dict`	Token usage from the upstream provider

`FileResult`¶

Field	Type	Description
`content`	`bytes`	Anonymized file bytes
`entity_count`	`int`	Total entities replaced

Links¶