Skip to content

Latest commit

 

History

History
150 lines (120 loc) · 8.16 KB

README.md

File metadata and controls

150 lines (120 loc) · 8.16 KB

op-tools

Tools for 1Password

Features

This tool is designed to provide an optimized workflow for reducing the amount of redundant information in your 1Password vaults. It operates on two main principles:

  1. Items that reference the same URL domain are most likely to be duplicates of each other.
  2. The more similar two items are, the easier it is for you to know what to do with them.

The first screen you'll encounter is a list of all of the "duplicate sets" that the tool has found in your vaults, grouped by domain, and ordered by "score", with the lowest-scored items at the top. When you click on one of these items, it'll open up the details page for that set.

On the details page for a set, you will only be shown those fields where the items contain different information from each other. The goal is to reduce the amount of information shown in a set until you feel comfortable taking one of two actions that will remove the entire set from your list of duplicates:

  1. Archive all but one of the items.
  2. Decide that the items shown actually represent different accounts you would like to keep, and tell the tool to ignore them in the future.

The easiest items to make decisions on will have very little information shown, because the items are largely similar to each other. For example, if only the title is shown, that means all other information is identical, and you can archive whichever one you'd like without fear that you're going to lose any vital information. More commonly, the URLs may differ slightly, as one of them may go to a login page and the other may be a deeplink to a profile page. Either way, if all other fields are identical, it's easy to decide which to archive, as it likely doesn't much matter.

When a duplicate set details page contains a lot of information, the easiest way to handle it is to use the "Copy to others" button to copy a particular field from one item to the same field on all the other items in the set. Working this way, you can reduce the amount of different information between items until you are left with a set of information you can more easily base decisions on.

Ultimately, if the difference between two items comes down to just the password, you're going to have to actually try logging into the website to figure out which password (if any) is actually the correct one. While I hope to make that a little easier from within the tool directly (see FR #7), for now, you'll want to jump into the 1Password desktop app to make further progress on that item.

Similarly, you will likely find some "duplicate sets" which are actually distinct accounts on the same service - perhaps you have two separate Google accounts, or your real IG page and a Finsta. In these cases, you'll want to take advantage of the "Ignore entire set" button. This will add an ignored_by_op_dedupe tag to each of the items in the set, and the tool will remove the set from the overall list view.

Installation

You'll need to have the 1Password command line tool installed on your system, and you'll need to have granted it permission to access your vaults from within the 1Password desktop app.

Once that's installed, you'll need to install Kivy, which is the GUI library on which we rely. The simplest way to do that is via pip with the following set of commands:

python3 -m pip install --upgrade pip setuptools virtualenv
python3 -m virtualenv kivy_venv

source kivy_venv/bin/activate
python3 -m pip install "kivy[base]"

Usage

App will use the Kivy GUI by default if available in the evironment, otherwise will fall back to the Tkinter GUI. I don't recommend the Tkinter GUI, and it's not likely to get a lot of love going forward.

source kivy_venv/bin/activate
./op_dedupe.py

Problems?

So far I've only tested any of this on a couple MacBooks running MacOS Ventura and Python 3.9.

If you run into any problems while using this tool, please do report them to me by filing a bug. Make sure to include your operating system version, what version of python you're using (get this by running python3 --version), the contents of any error messages that are printed to the terminal, and descriptions of what you did to get there, what you were expecting to happen, and what actually happened.

The more detail you can give me, the better, but don't include any password or username data, please.

Known Issues

You can see the full list of KIs here, but most importantly:

  • Issue #6: If you try to copy a list of URLs from one item to the others in a duplicate set, only the first URL in that list will be copied. This is a relative edge case, but be aware it exists. If this is blocking your workflow, let me know more about how and why in this bug.
  • Issue #5: A lot of your 1Password data ends up in an on-disk cache when you use this tool. I'll be building functionality into the app itself to delete the cache, but in the meantime, you can delete it yourself by running rm ./.op-cache/* from the same directory where you've been running the tool. There is now a button at the top of the list screen you can press to clear the cache before closing the app. If you're not planning on opening the app in a while, that's an easy way to clear out the on-disk cache. I'm still working on rewriting the cache to store less data going forward.

Why does this exist?

I was one of the many, many people who switched away from LastPass to 1Password after the huge breaches that LastPass suffered in 2022. When I made the switch, I chose to import into 1Password not only all my old LastPass entries, but also all the passwords I had saved in Chrome. This created a ton of duplicate entries, many of which were not perfect duplicates, as they varied in exactly what URL was associated with the entry, as well as sometimes varying in the stored usernames and passwords. Trying to pare down this list of duplicates at the same time I was trying to reset all my passwords was a huge pain, and 1Password didn't yet provide any built-in tools to assist with these tasks.

I created the first versions of this code in the passenger seat of a car driving down a rural Nevada highway, asking ChatGPT to draft up initial versions of the code and then trying them out and refining them until they did more or less what I wanted them to do. For that first version, my goal was to only use python built-ins, with the idea that this would make it easy for other people to use the same code without having to download a bunch of other dependencies. Unfortunately, the first time I tried to get my wife to run this code on her computer, I learned how awful the version of Tkinter is that is distributed by default with MacOS.

That initial version of the code using Tkinter is, as of this writing, still included in this project, and still technically usable, if you happen to have a good version of the Tkinter libraries running on your machine. However, I have since rewritten all the GUI code in Kivy and have made that new GUI the default if you have Kivy installed. I will likely remove the Tkinter version eventually, and I do highly recommend using the Kivy version. If you happen to prefer the Tkinter version and don't want me to kill it, I would like to know why.

The current version does a lot of on-disk caching. I initially did this because my internet connection was not stable while driving down the highway, but I later discovered that the 1Password API was slow enough that I really need to keep that cache around for the app to be reasonably performant. The first time you run the app, it's going to take a while to start up as it downloads all the necessary data. Subsequent times, it should only take a few seconds to start, unless you've cleared the cache manually. If you don't intend to run the app again for a while, I do recommend clearing the cache by following the instructions under the Known Issues section. I do hope/intend to make this process easier and clearer in the future.