Dev Note

How to Archive Multiple GitHub Repos

Published by Josh Briley on December 28, 2025

Dev Notes Disclaimer: Each artile in the Dev Notes section of my webiste may or may not be unfinished work. I don't always have time to write a full post. If you see something that looks like a half-baked idea, it probably is! As with anything you find on the internet, be sure to review and understand the script or code before running it, especially if you plan to modify it. Always back up any important data before making bulk changes to your repositories. If you have any questions or suggestions, feel free to reach out.

It's the end of the year and I figured I'd do a little Github pruning. I've created (and abandoned) over 100 repos over the years, so using the Github UI to archive all this mess would be awfully time-consuming. So I asked ChatGPT to help me write a quick bash script to do it for me.

For Mac folks, first install the Github CLI if you haven't already:

brew install gh

I'm not familiar with Windows, but Google tells me that you can install it with winget:

winget install --id GitHub.cli

Next, authenticate the CLI with your Github account:

gh auth login

Demo'ing my fizzbuzz and fibonacci repos isn't really necessary, but I didn't really want to delete these old relics. Nostalgia? Maybe. Hoarder? Definitely. I just wanted to archive the repositories and make them private.

Now create a bash script called archive-repos.sh.

#!/usr/bin/env bash
# Use the user's env to find bash
set -euo pipefail
# Exit on error, unset vars are errors, fail pipelines on first failure

# Read each repository identifier (owner/repo) from repos.txt
while IFS= read -r r; do
  [[ -z "$r" ]] && continue
  # Skips empty lines

  echo "Checking $r"            # Progress output

  # Query current archived status and visibility via GitHub CLI (JSON + jq selector)
  archived=$(gh repo view "$r" --json isArchived --jq '.isArchived')
  visibility=$(gh repo view "$r" --json visibility --jq '.visibility')

  # 1) Ensure PRIVATE (even if archived)
  if [ "$visibility" != "PRIVATE" ]; then
    # If archived, temporarily unarchive to allow changing visibility
    if [ "$archived" = "true" ]; then
      echo "  → Repo is archived and not private. Temporarily unarchiving to change visibility..."
      gh repo unarchive "$r" --yes
      archived_was_true="true"
    else
      archived_was_true="false"
    fi

    echo "  → Setting visibility to private"
    # Change visibility to private; the flag auto-confirms the consequences prompt
    gh repo edit "$r" --visibility private --accept-visibility-change-consequences

    # If it was archived before, re-archive now that visibility is updated
    if [ "$archived_was_true" = "true" ]; then
      echo "  → Re-archiving"
      gh repo archive "$r" --yes
    fi
  else
    echo "  → Already private"
  fi

  # 2) Ensure archived (after visibility is correct)
  if [ "$archived" = "false" ]; then
    echo "  → Archiving"
    gh repo archive "$r" --yes
  else
    echo "  → Already archived"
  fi

done < repos.txt

This script reads a list of repositories from a file called repos.txt. Create that file and add the repositories you want to archive, one per line, in the format owner/repo-name:

#repos.txt
your-github-username/repo1
your-github-username/repo2
your-github-username/repo3
# etc.

Make the script executable:

chmod +x archive-repos.sh

Finally, run the script:

./archive-repos.sh

A sample output might look like this:

Checking your-github-username/repo1
  → Setting visibility to private
  → Archiving
Checking your-github-username/repo2
  → Already private
  → Already archived
Checking your-github-username/repo3
  → Setting visibility to private
  → Archiving

And that's it! The script will go through each repository listed in repos.txt, make it private if it's not already, and then archive it. This should save you a lot of time compared to doing it manually through the GitHub UI.