Run a SQL query, attach the results as a CSV, render a PDF summary, and email the report. Designed for daily/weekly business reports — sales, signups, errors, anything queryable.

`database-snapshot-report`

Run a SQL query, render a PDF preview, attach the full CSV, email it. Designed for the most common business automation: "send me X every Monday morning."

Use case

Sales summary, signups by day, error counts, queue depth, anything you'd otherwise log into a database to check by hand. Replace 2 minutes of clicking around with one scheduled email.

This is a YAML workflow, not a skill — pure plumbing, no AI value.

Setup

Create a sql credential pointing at your database (vendor + server + login)
Create an email credential for SMTP delivery (host, user, password, fromAddress)
Edit the statement in workflow.yaml for your query
Set the to: address on the email task
Schedule (daily, weekly, monthly — whatever fits)

Pipeline

run-query — sql export → writes a CSV with column headers
format-preview — runtimeenv (bash + column) → fixed-width preview of top 100 rows
render-pdf — pdf generate → PDF from the preview text (Courier so columns stay aligned)
email-report — email send → both files attached, short body explains the contents

Why both PDF and CSV

The PDF is for skimming on a phone. The CSV is for opening in Excel/Sheets to sort, filter, or chart. Recipients pick the format that fits the moment.

Customization

Bigger reports — bump head -n 101 to show more rows in the PDF preview, or skip the PDF entirely and just attach the CSV
Multi-query reports — chain multiple run-query tasks each writing to a different CSV, attach all of them
Slack instead of email — replace the email task with a message task targeting a user; CSV/PDF won't attach to a chat message, so just embed the preview text in the body
Charts — for visual reports, pipe the CSV through gnuplot or a Python plot script in a runtimeenv task and attach the PNG

Notes

${INTOUCH_HOME}/reports/ must exist (mkdir at first use, or have the script create it)
The column -t formatter handles most CSVs but breaks on cells containing commas — for those, switch to a tab-delimited export (outputDelimiter: tab) and column -t -s$'\t'

Source

See workflow.yaml.