Skip to content

Home

pyatlan logo

The official Python SDK for Atlan ๐Ÿ’™

PyPI version Python versions License Downloads Build Status

Get started API Reference GitHub


Installation

pip install pyatlan
uv add pyatlan
git clone https://github.com/atlanhq/atlan-python.git
cd atlan-python
uv sync --group dev

Quick Start

from pyatlan.client import AtlanClient

# Initialize the client
client = AtlanClient(
    base_url="https://<your-tenant>.atlan.com",
    api_key="<your-api-key>",
)

# Search for assets
from pyatlan.model.fluent_search import FluentSearch
from pyatlan.model.assets import Table

response = client.asset.search(
    FluentSearch()
    .where(FluentSearch.asset_type(Table))
    .where(FluentSearch.active_assets())
    .page_size(10)
    .to_request()
)

for asset in response:
    print(asset.name, asset.qualified_name)

Features

  • Search & Discover


    Full-text and structured search across all your data assets โ€” tables, dashboards, glossary terms, and more.

  • Data Governance


    Manage access control, business policies, data contracts, lineage, and custom metadata at scale.

  • 260+ Asset Types


    First-class support for Snowflake, Databricks, dbt, Tableau, Looker, PowerBI, Kafka, S3, and many more.

  • Async Support


    All client methods available in both sync and async flavors via pyatlan.client.aio.

  • Type Safe


    Full py.typed marker, strict Pydantic v2 models, and comprehensive type stubs (.pyi).

  • Minimal Dependencies


    Only what you need: httpx, pydantic, authlib, msgspec, and a few more.


What's in this SDK?

Module Description
AtlanClient Main entry point โ€” all sub-clients live here
pyatlan.model.assets 260+ asset model classes (Table, Column, Dashboard, โ€ฆ)
pyatlan.model.search IndexSearchRequest, Bool, Term, and other DSL builders
pyatlan.model.fluent_search Fluent builder API for composing searches
pyatlan.cache In-memory caches for tags, metadata, groups, roles, users
pyatlan.events Lambda / event-handler base classes
pyatlan.model.typedef TypeDef management (custom metadata, enums, structs)

Project Stats

  • ๐Ÿ Python: 3.9 ยท 3.10 ยท 3.11 ยท 3.12 ยท 3.13
  • ๐Ÿ“ฆ Stability: Production-ready (5 - Production/Stable)
  • โš– License: Apache 2.0
  • ๐Ÿš€ Async: Native async/await support

Contributing

# Fork & clone
git clone https://github.com/atlanhq/atlan-python.git
cd atlan-python

# Install dev dependencies
uv sync --group dev

# Run tests
uv run pytest tests/unit

# Run all QA checks (lint, type-check, format)
uv run ./qa-checks

See CONTRIBUTING.md on GitHub.