Choose Your Path

You can explore in the browser or query from your own environment.

Log in at oso.xyz, go to Notebooks, and click New Notebook
The setup_pyoso cell is auto-generated — you're already connected to the data warehouse
Add a new cell and paste a SQL query. Here's one to try — largest ecosystems by repo count:

SELECT
  e.name AS ecosystem_name,
  COUNT(DISTINCT er.repo_id) AS repo_count
FROM stg_opendevdata__ecosystems e
JOIN stg_opendevdata__ecosystems_repos_recursive er
  ON e.id = er.ecosystem_id
GROUP BY e.name
ORDER BY repo_count DESC
LIMIT 15

For more notebook patterns, see the Agent Guide's notebook creation guide.

1. Get an API key — Go to Settings > API Keys and create a new key. 2. Set your environment variable:

export OSO_API_KEY=your_api_key_here

3. Choose your workflow: Option A: marimo notebooks (recommended)

uv add pyoso marimo
uv run marimo edit notebook.py

Queries use mo.sql() with the pyoso database connection:

import pyoso
import marimo as mo
pyoso_db_conn = pyoso.Client().dbapi_connection()
df = mo.sql("SELECT * FROM oso.projects_v1 LIMIT 10", engine=pyoso_db_conn)

Option B: Any Python environment

pip install pyoso

from pyoso import Client
client = Client()  # reads OSO_API_KEY from environment
df = client.to_pandas("SELECT * FROM oso.projects_v1 LIMIT 10")