Byline

What leaves your machine

What Byline collects, exactly

Byline only sees what you connect. Nothing is scanned, discovered, or collected by default.

GitHub connection. We read your own pull requests, issues, reviews, and commits with read-only permissions. We never read your organization's data. We store titles, bodies, diff statistics, and links.

The collector (your machine). You add repos one at a time. Each repo has a tier you choose:

TierWhat leaves your machineWhat never leaves
fullCommit messages, file paths, diff statisticsYour code
titlesFirst line of each commit message, diff statisticsBodies, file paths, your code
countsWeekly totals: commit count, lines changedAll text, all names, your code

Your code never leaves your machine at any tier. Diff statistics are numbers, not content.

Repos with a work remote default to the titles tier. Before the first upload from any repo, the collector shows you the exact payload and waits for your yes.

Secret patterns (keys, tokens, credentials) are scrubbed from every outbound field at every tier, always.

A caution about work repos

Your employment agreement may restrict what you can share about your work, even privately. Commit messages can contain confidential names and details. Before you connect a work repo:

  1. Prefer the titles or counts tier. When unsure, choose counts — numbers only.
  2. Read the first-sync preview. It is the exact payload.
  3. If your agreement is strict or unclear, ask your employer before connecting the repo.

Byline gives you controls and previews. It cannot judge your agreement for you. This page is not legal advice.

What Byline never does

Removing your data

byline remove <repo> plus one click removes a repo's data from your ledger. Deleting an entry removes it from every surface. A lapsed subscription freezes synthesis; it never touches your data or your public page.

What leaves your machine — Byline