informational*.sqldump.sqlbackup.sql

Database dumps (.sql)

SQL dumps can contain production data — including personal information — that should not be indexed by an AI assistant.

This week: 2 repos in the scanned fleet exposed this category (2 matches).See Radar #4 →

What it is

A database dump is an export of schema and rows, used for backups or local seeding. Real dumps often include user records, emails, and other personal or sensitive data.

Why it matters for AI context

When a dump sits in the repo, an assistant can read its rows as context, pulling personal data into prompts sent to model providers. This can create privacy and compliance issues.

How to keep it out of AI tools

  1. Keep production dumps out of the repo; use anonymized seed data for development.
  2. Add *.sql dumps (or a dedicated dumps/ folder) to .gitignore.
  3. Add the same to your AI ignore file.
  4. Scrub personal data from any sample data you do commit.
Add to your ignore file
*.sql
dumps/
*.dump

Ignore files that exclude this

FAQ

Are schema-only .sql files a problem?
Schema files are usually fine to commit. The risk is dumps that include real rows of personal or sensitive data.

Check your own project

158 of 499 repos this week exposed sensitive paths to AI tools. Check yours in seconds:

Run the same local check with Offsend CLI:

brew install --cask offsend/tap/offsend-cli
offsend show
offsend prepare --dry-run

Everything runs locally. No file contents are uploaded.

Scan with Offsend CheckLearn more about Offsend →