---
id: steel-sdk
version: "0.19.0"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# steel-sdk — The official Python library for the steel API

License: permissive · Maintenance: active · Downloads: 154.5K/mo

## What it is and what it does

Steel SDK is the official Python client library for the Steel REST API, generated using Stainless. It provides both synchronous and asynchronous clients powered by httpx, with full type annotations for all request parameters and response fields using Pydantic models and TypedDicts. The library handles common REST patterns including auto-pagination, file uploads, nested parameters, and comprehensive error handling with automatic retries for transient failures.

Developers use it to integrate Steel API functionality into Python applications without manually constructing HTTP requests. The library supports modern Python versions (3.9 through 3.14), offers editor autocomplete and type checking, and can optionally use aiohttp as an HTTP backend for improved async concurrency. Configuration is simple—credentials are typically loaded from environment variables—and the API surface mirrors the REST documentation.

Use it for:

- Build automation tools that create and manage Steel browser sessions programmatically.
- Integrate Steel's browser automation into async web applications or event-driven systems.
- Batch process large lists of sessions with auto-pagination without manual page handling.
- Upload files to Steel via the typed file upload interface with automatic content-type handling.
- Handle API errors gracefully with specific exception types (RateLimitError, AuthenticationError, etc.).

## Worth the install?

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

Provides a typed Python client for the Steel REST API with both synchronous and asynchronous interfaces, auto-pagination, and comprehensive error handling.

Yes. The package has low install friction, active maintenance, no known vulnerabilities, permissive licensing, and comprehensive type support. Install it if you need to call the Steel API from Python; the typed interface and async support make it substantially easier than raw HTTP requests. Verify that your use case aligns with Steel's service offerings before committing.

## Install

pip install steel-sdk
uv add steel-sdk
poetry add steel-sdk

## Installing steel-sdk

Before you install: Low friction install with six standard dependencies (anyio, distro, httpx, pydantic, sniffio, typing-extensions). Actively maintained with recent commits and no known vulnerabilities.

License in practice: Apache-2.0 is permissive; you can use, modify, and distribute this library freely in commercial and private projects with minimal restrictions.

Quickstart:

pip install steel-sdk

from steel import Steel

client = Steel()
session = client.sessions.create(
    api_timeout=20000,
    use_proxy=True,
)
print(session.id)

Requires Python 3.9 or later; Steel API key must be set via STEEL_API_KEY environment variable or passed to client constructor.

Verify before relying:

- Whether Steel API service itself is production-ready and has documented SLAs.
- Rate limiting behavior and retry strategy details beyond the documented 2 default retries.
- Performance characteristics under high concurrency with aiohttp backend vs. default httpx.

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 154.5K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

steel api client python, python rest api client library, typed http client sdk, async sync api wrapper, pydantic api client, httpx based sdk, auto-paginating api client, api-client, async-support, type-safe

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