Skip to main content
Jmail No install required — uses PEP 723 inline dependencies with uv. 
# Show first 5 emails
uv run clients/python/jmail.py emails --head 5

# Network-only view (no body text, much smaller)
uv run clients/python/jmail.py emails --slim --head 10

# List all dataset URLs
uv run clients/python/jmail.py urls
Or use as a library: 
from jmail import JmailClient

client = JmailClient()
df = client.emails(slim=True)
print(df.head())

​
DuckDB (zero download)

Query Parquet files directly over HTTP — nothing to install beyond DuckDB: 
SELECT sender, COUNT(*) as n
FROM read_parquet('https://data.jmail.world/v1/emails-slim.parquet')
GROUP BY sender ORDER BY n DESC LIMIT 20;

​
Direct Download

Grab any dataset directly: 
# Parquet
curl -O https://data.jmail.world/v1/emails-slim.parquet

# NDJSON (gzipped)
curl -O https://data.jmail.world/v1/emails-slim.ndjson.gz

​
What’s Next?

Browse Datasets

See all available datasets and their schemas.

Python Client Reference

Full client documentation with caching, options, and CLI.