Twitter Archive Site

Turn the archive you downloaded from Twitter/X into a static website you can publish anywhere — keeping your private data private.

Supported format: the Twitter archive as exported in April 2024.

Live example

See what you get: twitter-archive.marcomaroni.it is the author's own archive, generated with this tool from the original Twitter export and published on GitHub Pages (source).

What you need

• Node.js 18 or newer
• Your Twitter archive (.zip), requested from Twitter/X under Settings → Your account → Download an archive of your data

Create your site

1. Download this project (green Code button → Download ZIP, then extract it) or clone it with git.

2. Copy your archive .zip into the archive folder. If Twitter split it into several zips, copy all of them.

3. Open a terminal in the project folder and run:

   npm install
   npm run build
   

4. Answer a few questions. Pressing Enter keeps the safe default.

5. Double-click site/index.html to check the result in your browser.

6. Publish the site folder (see Publishing).

Your answers are saved in config.json: the next npm run build asks nothing.

The site folder is deleted and recreated on every build: do not add files to it by hand. Put them in the extra folder instead: everything in it (except its README) is copied into site on each build.

Command	What it does
npm run build	Builds the site (asks questions only the first time)
npm run build -- --reconfigure	Asks the questions again
npm run build -- --yes	Never asks; uses config.json or the defaults
npm run build -- --debug	Shows technical details when something goes wrong

If you stop a build (Ctrl+C or by closing the terminal), the temporary copy of your archive is removed; anything left behind is removed at the start of the next run.

Git Bash on Windows: its terminal (mintty) does not show the setup questions. Run npm run build from PowerShell, Command Prompt or Windows Terminal instead, or edit config.json by hand, or run npm run build -- --reconfigure from one of those terminals.

What gets published — and what doesn't

Published:

• Your own tweets and the replies you wrote in your own threads
• Their photos and videos (with location and camera data removed from images)
• Your name, username, bio, avatar and header as shown on your profile
• Retweets and replies to other users only if you choose so (default: no)

Never published, whatever you choose:

• Direct messages, email address, phone number, IP addresses
• Followers, following, likes, blocks, mutes, lists, ads data
• Deleted tweets
• The device you tweeted from and the location attached to tweets

The tool reads only the files it needs from the archive. If one of your tweets contains your own email, phone number or IP address, that tweet is hidden; if your bio or the website you chose contains one, that field is left empty. The summary at the end tells how many tweets were hidden (never the values). As a last safety net, after building it searches the site for your email, phone number and IP addresses: if it finds any, it deletes the site and stops.

Mentions of other people (@name) inside your tweets are kept as they are, because they were already public.

By default the site asks search engines not to index it. You can change this during the setup.

Publishing

The site folder is a plain static website: upload it anywhere.

• GitHub Pages: create a new repository, put the contents of site in it, then Settings → Pages → Deploy from a branch. GitHub rejects files over 100 MB and recommends sites under 1 GB: the build warns you if you exceed them. For a custom domain, create a file extra/CNAME containing only your domain (e.g. tweets.example.org): it is copied into site on every build.
• Netlify: drag and drop the site folder on https://app.netlify.com/drop.
• Your own server: copy the contents of site to the web root.

Never publish the archive folder or config.json.

Development

npm test

Tests run on a small fake archive in test/helpers/fixture.js, never on real data.

Author

Created by Marco Maroni.