eng-handle-customer-data

Files (1)

• SKILL.md

SKILL.md (first lines)

> Generated file — do not edit.
> Persona: engineer (Software engineer)
> Pack:    engineer-pack v0.2.0
> Source:  skill `handle-customer-data` @ 3d7e0fd
> Edit the source skill, not this copy — this one is overwritten on every build.

# Handling customer data

## Boundaries that apply to this skill

**Use the least-sensitive environment that can reproduce the issue.** Start with synthetic fixtures
or approved non-production data. Do not copy customer records, production database rows, or
unredacted logs into source control, tickets, prompts, or local test data.


## The line

Personal data stays in the system that holds it. Aggregate it, count it, join it — but an extract
containing identifiable records leaving that system needs a named owner who approved it, recorded
before the extract is made rather than after.

## Working without it

Most questions that seem to need identifiable records do not:

- **Counting** — aggregate in the query, extract only the aggregate.
- **Sampling for a demo** — generate synthetic records that share the shape and the edge cases. A
  synthetic set you designed is usually a *better* demo, because you control the edge cases.
- **Debugging a specific record** — work in the system, quote the field that matters, not the row.

## What to do when asked to cross it

Reproduce the behavior with synthetic fixtures or approved non-production data. Redact logs at the
source and keep customer rows out of source control, tickets, prompts, and local test databases.

From aws-samples/sample-skillforge, Apache-2.0.

More from aws-samples/sample-skillforge