Skill · Code & Agents
Data Pipeline Builder
From hand-merged exports to a re-runnable pipeline whose report totals tie back to the source files, with checks that stop bad data.
When to use
- You combine exports (CSV, spreadsheets, API pulls, database tables) by hand on a schedule.
- Two reports disagree, or a report disagrees with the source system, and nobody knows why.
- You are starting a small analytics or ETL job and want a sound structure before the code grows.
When not to use
- A one-off analysis of one clean file. Just analyse it.
- Streaming at large scale, or a choice of data-platform vendor. That needs a specialist architecture review.
Example
User: Every Monday I merge a Shopify orders CSV and a Stripe payouts CSV in Excel to get weekly net revenue. I use Python with pandas.
Result (excerpt):
Cleaning rules
Stripe: amounts arrive in cents → divide by 100; fee is a separate column. Shopify: refunds arrive as extra rows with a negative total and the original order name.
| edge case | handling |
| order with no matching payout | quarantine to unmatched.csv, warn |
| same Shopify file loaded twice | the week's output is replaced, never appended |
| payout dated in the future | block the run |
Validation gates
| gate | check | on failure |
| raw | columns Name, Total, Created at present | block (Shopify renamed a column) |
| join | at least 98% of orders matched | warn below 98%, block below 90% |
Reconciliation
| net revenue | Shopify 18,420.50 | quarantined 310.00 | report 18,110.50 | yes |
Code
def load_raw(path): ... # read, log row count, add source_file and source_row
def check_raw(df, expected): ... # gate 1: schema and dates
def clean(orders, payouts): ... # cents to EUR, refunds, join on order name
def write_week(df, week): ... # replaces output/<week>.csvIllustrative example written by Meevra; your model's wording will differ.
What is inside
The full skill has these sections. The ones marked “in the pack” come with the download.
- When to use on this page
- When not to use on this page
- Inputs to ask for in the pack
- Workflow in the pack
- Output format in the pack
- Example on this page
- Model notes (Oct 2026) in the pack
- Troubleshooting in the pack
Get Data Pipeline Builder
The full skill is a SKILL.md folder you install once and reuse: the workflow, the checklist and the answer format written out, with notes for each model. It comes in the Code & Agents pack (5 prompts · 15 skills · 1 free), with a QUICKSTART, setup notes in every item and 12 months of updates.
All-Access 39 € vs 7 packs × 12 € = 84 € · less than buying 4 or more packs one by one
Stuck? Email support@meevra.com — we'll help you set it up.