---
id: sift
version: "6.0.0"
license: The MIT License Copyright (c) 2013-2016 Sift Science (https://siftscience.com) Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated… (full text in the JSON record)
license_treatment: permissive
maintenance: aging
---

# Sift — Python bindings for Sift Science's API

License: permissive · Maintenance: aging · Downloads: 295.3K/mo

## What it is and what it does

Sift is a Python client for Sift Science's fraud detection platform. It wraps the Events, Labels, Score, and Decisions APIs, allowing you to send transaction and user activity events to Sift for real-time risk assessment, retrieve fraud scores and percentiles, label users as fraudulent or legitimate, and manage verification workflows including OTP generation and validation.

The library is designed for integration into payment, marketplace, and authentication systems where you need to detect and prevent fraud. It handles API communication, response parsing, and error handling through a simple client interface. No external runtime dependencies are required beyond Python itself.

Use it for:

- Track e-commerce transactions and user events to Sift for real-time fraud scoring and abuse detection.
- Retrieve risk scores for users or orders to make accept/reject decisions in payment flows.
- Label users as fraudulent or legitimate based on manual review or chargeback events.
- Implement OTP-based verification workflows for high-risk login or transaction events.
- Query workflow execution status and latest decisions for users, orders, sessions, or content.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Python client library for Sift Science's fraud detection and abuse prevention APIs, providing methods to track events, request risk scores, label users, and manage verification workflows.

Yes, if you use Sift Science for fraud detection. The library is stable and supports current Python versions, but high install friction and aging maintenance (466 days since last release) suggest checking whether Sift has released a newer version or recommends an alternative client before adopting. No known vulnerabilities.

## Install

pip install sift
uv add sift
poetry add sift

## Installing Sift

Before you install: High install friction with no runtime dependencies. Last release was 466 days ago; repository is not archived but shows aging maintenance. Supports Python 3.8 through 3.13 and is marked Production/Stable.

License in practice: MIT License (permissive). You may use, modify, and distribute this package freely in commercial or private projects, provided you retain the original copyright notice and license text.

Quickstart:

pip install Sift

import sift

client = sift.Client(api_key='your_key', account_id='your_account')
response = client.track('$transaction', {'$user_id': '23056', '$amount': 15230000})
if response.is_ok():
    print('Event tracked')

Requires a valid Sift Science API key and account ID to authenticate requests.

Verify before relying:

- Whether the high install friction relates to compiled extensions, network checks during installation, or other build-time requirements not evident from the package metadata.

## Package facts

- License: The MIT License Copyright (c) 2013-2016 Sift Science (https://siftscience.com) Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated… (full text in the JSON record) (permissive)
- Python support: supports_current
- Install friction: high
- Maintenance: aging
- Downloads: 295.3K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

fraud detection api client, sift science python bindings, abuse prevention sdk, risk scoring api, event tracking fraud, fraud-detection, api-client, sift-science

[View on SkillFed](https://skillfed.io/packages/sift) · [View on PyPI](https://pypi.org/project/sift/)