Clean Up Your Inbox In Minutes
Newsletter management, AI automation, and email analytics.
Inbox Zero is the open-source email app that puts you back in control of your inbox.
This is probably a symptom of my severe boredom, but I tried to find the tweeted testimonials on Twitter, and it seems quite a few of the testimonials do not correspond to real tweets.
The testimonials by "@james", "@yoni", "@ktyr" are pretty obviously not real just from their implausibly short Twitter handles. The actual Twitter handles have different profile pics that simply don't match the ones displayed on the landing page. For example @ktyr is not an Indian male but an Asian girl who hasn't tweeted since 2012.
@prem_saini1 also is not the same person.
@emarky is the same person, but he doesn't seem to have tweeted the testimonial.
The ProductHunt link at the top suggests perhaps that is where the testimonials are from. It's not clear why you think they're meant to be tweets. A quick scan of the Product Hunt page quickly found at least one that has a reply saying they'll add it to the testimonials if ok.
Nothing on the page says they are all reviews from producthunt, nor that the testimonials are all from public sources. I get being skeptical; my default assumption tends to be that testimonials are not to be trusted, but this seems excessive.
Here’s a tip: if you modify reviews for “privacy reasons”, there is literally nothing separating them from a fake review.
This is not how you build trust. If someone accepts being listed as a review there is no privacy concern, if they do not they shouldn’t be there at all.
I have a serious issue with products that claim to be email solutions in every corner of their marketing and then, buried in the FAQ, say that they’re GMAIL only. This is bait and switch for my time. I have multiple Gmail accounts, was considering the lifetime sub but I’ll pass.
"We only support Gmail and Gsuite email accounts today"
Nice work here but here's where innovation could start with another smart email client in the crowded market. It seems they are all "gmail now and outlook on the roadmap" I know these are common email suites but why not friendly and open protocols like Fastmail's JMAP?
Gmail is also precisely the product that needs this extra functionality the least.
It’s a lot harder to figure out if you have product market fit when you’re choosing a starting point that significantly reduces your addressable market.
Do GMail and Outlook (and Exchange?) support JMAP?
I'm all for open standards, but as a sibling put it, when you're starting out, you may not want to limit your market size, unless you know you're addressing some kind of niche to begin with.
Most apps target Gmail because most users use Gmail.
Supporting other email providers would be nice to have.
It's open source so maybe someone else will contribute support for other emails before I get to it. It's not the first item on my roadmap. I'd like to perfect the current user experience.
Tbh it would make more sense for other providers like Fastmail to provide a Gmail-compatible API since that seems to be the defacto standard. They could open source the Gmail to JMAP translation layer if they want to improve interoperability.
“By continuing you agree to allow Inbox Zero to send your emails to OpenAI for processing.”
Yeah, unfortunately this kills it for me. Handing my whole personal inbox over to OpenAI seems insane.
I'm not sure if that's already exposed as an option, but if you're happy to run this app locally, then LiteLlm exposes local models using the OpenAI API. That means you only need to change the endpoint, tell LiteLlm to use for example Mistral and it should "just work". Even if not implemented yet, it would be a trivial change.
Trivial change, except for also having to change all the prompts to work well with Mistral. LLMs are not equal, you have to ask each one "nicely" in a different way.
You can change the prompts if you want better results. But you don't have to do it. I've replaced GPT endpoint with LiteLlm a few times already and had to adjust the prompt only once. If I can't see the difference, why bother?
The project uses the ai npm package from Vercel which does interface with a lot of different LLMs.
But I imagine there is a little work for it to just work with any provider.
I think it would be really cool to get it running with something like Mistral in the browser: https://webllm.mlc.ai/
If you are looking for a simple solution for inbox zero for gmail - quickly classify emails as archive, follow-up, read through, take a look at my hacky attempt.
It is inspired by vim. It downloads all the email snippets and sorts them in descending order by sender and date.
Then you use “j”, “k” to navigate through the emails. You can press “a” for archive, “f” for follow up, and “r” for read through. You can also press “p” to navigate to next email prefix (the stuff before the @) and “d” to navigate to the next domain. Like vim you can combine keystrokes. “ad” will archive all the emails from the present position to the next domain. Since these are sorted by reverse date, often you can look at recent dates and archive the rest.
All operations happen in memory so they are instantaneous. You the press “w” to actually move the messages on the server as a batch operation.
It is really useful if you are getting started or are falling behind and need to quickly classify a few thousand emails.
It is a single 500 line Rust program so you can hack on it to change it to how you like it.
https://github.com/neomutt/neomutt is a mature project with similar functionality in its arsenal via tagging and vim bindings, not limited to gmail either. You can also write emails in vim/$EDITOR fluidly.
I highly doubt he’s involved with this, and he would probably be pretty annoyed with the name being used.
My understanding is that he’s grown to be a bit resentful of how the term has been carried away and used, and he cringes/dislikes software that “misuse” the term.
The AI-powered rules thing is revolutionary. I can't even begin to imagine what it can do, but I'd rather it didn't just do it automatically, but rather put it into a queue for me to approve. I'd love an AI assistant to write draft replies automatically for me to edit and send out.
My budgeting app (Monarch) has a system with a lot of parallels. It tries to automatically profile your transactions into a sensible category based on the statement text, and also allows you to set up persistent rules on top to categories it may have messed up. It still has a “we couldn’t categorize this, please help” list it gives you, but I find its recommendations fairly good and I get maybe one transaction a month it can’t figure out.
The kind that runs locally, or that is run by my trusted (by me, by my employer, ...) email provider. Email is just too valuable to be exposing to extra third parties.
Phones have made this more inconvenient because it's hard to make 'the app that runs locally' run whenever 'the app that views my email is running'. That means running locally is a big constraint on the UX, for example it probably can't reliably sort emails before my phone picks them up. It's simply the case however that UXes that don't work with local apps aren't viable products to sell to me.
Open source not running locally is closed source, for all intents and purposes.
A GNU/Linux box supposedly loaded with nothing but software libre is closed source, if it's remote to you and you don't own it. If you send your data to it to be processed, who knows what happens to it.
I hear what you’re saying, and I agree that running something locally is appreciably different than using something hosted elsewhere.
But I think it’s critical to be precise about this. Open source is still open source, and this designation is still meaningful even if an open source product provides a hosted version. The downsides of letting someone else host something are independent of the the open/closed source status.
Yeah, and I briefly considered that, but in truth I can't be bothered to run software stack of this complexity for my personal use for this. It's designed as a mutli-user web app, and that shows in the number of different services used in the backend. I like the idea of an app that sorts my email better, but not that much.
I would say I don't consider this a criticism of the product - they're clearly selling a hosted product that they were kind enough to open source, not making a product for end users to run that they happen to sell a hosted version of. It being inconvenient to self host doesn't reflect poorly on them.
It just needs to be proven and matured a bit more. It doesn't give the impression that much time has been spent building it. There is no indication that a third party has reviewed it for potential problems.
I don't think that's too high a standard. There is a Gmail email client, Mimestream, that you have to fully trust to use the Gmail API responsibly. It's gone through an extensive security review from Google's side to be allowed to use that API at a high volume. It was also in beta for a couple years and fixed a ton of bugs and edge cases during that period. There is evidence that a small team was hired which put more eyes and specialized minds on issues.
Mimestream has well exceeded the threshold to be trusted. I think this app could get to the threshold sometime next year if they work and communicate the right way. I kind of doubt they will spend that effort.
That's my opinion as an actual potential user (unlike those wanting it to be made of entirely locally run services, who weren't going to use it anyway.)
"It's gone through an extensive security review from Google's side to be allowed to use that API at a high volume."
Every app that has access to the Gmail API has been through this process including Inbox Zero. But I agree that you can have more trust in products that have been on the market for longer.
The safest way to run the app is to self-host but I agree that's a hassle for most to do.
I'm not the OP, but I would only trust an app like this if it didn't require any external services whatsoever. If every component can be self hosted + they add support for a local maildir or maybe IMAP, then it would be fun to try self hosting an instance of this app, but without that I wouldn't have any interest.
I mean it says you can self host it yourself... and its on github and opensource, and if it's using openai api, you can always use a self hosted api that's compatible with openai api to run mistral or llama to run it all locally.
Yes, you're right, but I don't know if it's possible to replace the tinybird component with anything local, perhaps there's a convenient alternative I'm not aware of (I'd never heard of tinybird before seeing it in the README's list of external services). Either way, the documentation doesn't have a lot of information about a local-only use case, so I can only imagine it would be inconvenient. If they add support for self hosted mail sources in the future, it may still be worth trying though.
Yes, atm it relies on Tinybird. But the level of trust you need for that is different. And the data in Tinybird is encrypted so even if you don't trust them to run a Clickhouse database, if you set an encryption key they won't be able to do much with your data.
We are planning to support a version that doesn't rely on Tinybird at all but stores all data in your browser using indexeddb.
Weird that it’s only offered to Gmail or Gsuite users. Email is supposed to be provider agnostic. I guess this app is using Gmail/gsuite proprietary apis to do its magic.
I've been using Mailstrom service on and off for years for bulk classifying my mail and some automated rules, but it seems like they haven't added anything new in years (I think they are focusing in some sort of iPhone app now) and I've tried to contact them to ask a couple things and they don't answer my emails.
So I've been thinking to make some sort of similar app to go through my mail metadata and make grouping and rules... "Delete this emails after a week, move all these others just when you receive them, send this others straight to the bin" with some extra smartness added.
Never thought I'd see my old employer mentioned on HN! When I left in the mid-2010s, they were just getting the mobile app off the ground. The first iteration was just trying to replicate some of the web app functionality with a new UI but I haven't checked in on it in years.
I have to make a correction now: I've been checking my mail and I've found that Leslie answered me but the message ended on my spam folder (hah!). The answer: They didn't have plans for the functionality I asked for. I asked for the possibility of deleting or moving certain emails after certain number of days (for example, for newsletters, billing emails, notices, and things like that).
As of today, Mailstrom's functionality seems to be the same as 5 or more years ago. Which is not bad-bad, but not enough for me. It seems like the company is focusing on an iPhone app called Chuck [0] that's kind of similar but plus AI and some other stuff. I'll go build my own Lunar Lander.
I fully spec'd out "On the Way" with a buddy of mine. Picking up beer close to your destination? Gas close to your departure point? The inverse? We would have you covered.
We had a path with Google Maps API, and I was convinced that monetization was at least possible enough to get Real Life VC funding.
In any case...this looks like a couple features. Sorry... :(
It was a feature. Google Maps had it implemented to 75% of what we'd spec'd within 9-12 months.
It's not like we'd actually tried, of course. We had full time jobs, for God's sake! But it became abundantly clear to me in that timeframe that FEATURE-sized ideas weren't gonna be viable. The Big Boy Ad Companies were gonna be burning those down for the next few months/years/forevers.
I've tried running a few open source apps like this locally, sometimes it's as simple as standing up a Mock OpenAI endpoint but if they use fancy features like embeddings endpoints then it can be more complicated. The quality is getting better, but it's always not-the-same.
My interpretation of inbox zero (and the only approach that keeps me sane):
- use rules to move automated emails to sub-folders
- anything that requires work or a response goes to an "unhandled" folder
- anything else in inbox goes to archive after I've read it
I've written a locally running app that is a lightweight version of this. You can download all your emails, it loads it into sqlite, and you can browse and sort by aggregate email counts (how many emails each sender sent). Once you select a sender, you can archive all of the emails it sent you.
I'd have loved to have made it a full fleshed app like this. Kudos!
I stopped doing Inbox Zero when I realized email search (O365 and Gmail in my case) is sufficient for my needs. I just don't have the energy to sort all my email, even with AI help. All I do lately is the occasional pass to delete large attachments and newsletters.
This was my paradigm through the last years and I'm always promoting it to my colleagues. What for should invest something in folder structures (representing only one of many possible criterias)? Reliable automation for this task find its limits when subject-lines are not written in in a formally complete an consistent manner. And of course since Googlemail preached "never delete" we all stick to some email hoarding attitute (better having it than needing it, especially for long running projects there are some arguments). This way I left a 90 GB Mailbox at the company I worked for before. :-)
But such unstructured bunch of emails is hard to manage, if any subset is to be handled over to another person or to be archived in some other environment. Furthermore size limitations for the inbox will force you to get rid of something earlier or later. And there is no fun in going through thousands of emails from years ago. Therefor I recently started to at least using a rough structure of folders for archiving (all mounted as sub-folders of Inbox and therefore searchable at a glance).
Some categories (ToDo, Waiting for result etc.) I'm realizing by categories in Outlook. This is quite handy to follow-up as search-queries for those categories and more can be pinned as virtual folders in Outlook.
However, I wish there were more tools in Outlook for categorizing emails while leaving them all in the same place. I'm not sure whether I ever will become an exemplary user of Inbox Zero or even reach this state.
Agree, you shouldn't overdo your email. Many people waste time organising for no benefit. The goal of the Inbox Zero app is to make you more efficient. Not that you spend even more time cleaning your email.
GDPR requires newsletters (not talking about spam) to include unsubscribe links (or similar) in the mails they send. Those are most of the times in the mail body and also sometimes in the mail headers.
But in reality, don’t a proportion of these unsubscribe links do what the previous poster suggested: link you to a proprietary system that requires you to log in to ‘update your mail preferences’?
I recently had to solve a captcha in order to unsubscribe from something!
The unsubscribe link included a cookie in the URL which identified what is being unsusbscribed. Only the e-mail recipient would reasonably follow that link. And yet captcha in your face.
Kind of reminds me of the old POPFile (and the IMAP extension) I used to use years ago. Email comes in, is classified and sorted into the appropriate folder.
I had always intended to add labels to it but never got around to learning Pearl to do it.
Popfile is a per folder bayesian auto classification system, procmail's more of full fat server mail filter system. You could get procmail to do the same thing but it would require a bit of work and certainly isn't an out of the box sort of thing.
ATM it's per account, but if we'll make it easier to add multiple accounts in the future. If you're buying to day then feel free to email elie at getinboxzero to help out.
For me, the hardest part about inbox-zero is that I use my inbox as a todo list. And a lot (most?) of my job involves reading an E-mail where Alice has a problem and needs me to E-mail Bob to get them to do something. So now I have two things in my todo list: 1. hound Bob until he finally reads his E-mail and does the thing, and 2. follow up with Alice to let her know the thing is done. I haven't yet found any great way besides email-as-todo list to manage this. Multiply this by 15-30 things at a time, and I end up with at least 30-60 things in my inbox.
Gmail's "stars" are an OK solution, but I tend to forget to unstar things when they're done, so I need to periodically do star-cleanups. And copy pasting tasks from my inbox into some other specialized todo tool has never really been worth the trouble. The todo apps just aren't integrated enough with the rest of my E-mail based workflow. And there's never a point where I'm not waiting on someone so I never really get to zero.
> I haven't yet found any great way besides email-as-todo list to manage this.
Which is fine. Put them in a To-Do folder. The original formulation of Inbox Zero suggested three actions:
1. Take care of the e-mail right now. Then archive it.
2. If you can take an action, but not right now, move it to “Follow Up”.
3. If you need someone else to take an action, move it to “Hold”.
It’s _Inbox_ Zero, not e-mail zero. Get the inbox – the place for unprocessed things – empty. Get it empty by processing the things. Then go do something other than e-mail.
This reads a bit like a microcosm of David Allen's "Getting Things Done" system.
Unprocessed emails = "Collection Bucket"
Follow Up folder = "Projects" list
Hold folder = "Waiting For" list
Except with GTD you do this with not just emails but also text messages, DM's, snail mail, personal lists you've jotted down, etc. It's a life organization system.
IMO, the above entity 'types' are mutually exclusive and collectively exhaustive categories (any concept or thing you can think of fits neatly into only one, and they should never be conflated or confused with each other). I think it is important to keep their identities very clear and very separate.
I.e., Task =/= Message and I think a system that recognises that is usually better (but, each to their own).
Sorry to clarify, an email is only ever a Message. It should not be treated as a Task (IMO).
Sender = Contact
Email = Message
Recipient = Contact
Email may contain an invite to a meeting = Event
Meeting room, coordinates, address or whatever of the meeting = attach Location
Email may ask you to do something before meeting = derive a Task
Audit trail item of you archiving the email = Log
Email Template that you use in your reply = Blueprint*
Rule to send an email at a certain time = Blueprint*
* Blueprints are just nested conditionals if you think about it... The subtype 'Templates' are nested conditionals without* a trigger IF (it's just a 'do', not an 'if then do'), and the subtype 'Rules' are nested conditionals with* a trigger IF (if else etc. then do)
This categorization seems to be based on one dimensional thinking (not in a bad way).
If you bring in more dimensions you end up with something like a taxonomy where an email contains a message and maybe also a todo or something else and may have more attributes around those identities. Then you can filter by those.
I do not use that personally but I see there is no reason to define an email as a todo item.
Also, as an email is just a storage for data, you could (mis)use it as a storage backend for a file system (see fuse) or backup solution.
You end up with a graph. Entities are the nodes, and the entity-relationships are the edges.
An email is always a message, nothing more. Tasks, events etc. can be derived from it, but the email itself is not a task or event. This isn't arbitrary, these boundaries are clean as heck (again of course, with the 'IMO' caveat. I don't care to convert anyone here).
Very interesting model — what do you actually do with this conception? Do you have software that helps you reason about your inbox/rest of life this way?
At the moment, personal stack only. React native web local + react query + async storage + supabase + etc.
If I pull up a contact 'MyClient' I can see every entity that's ever been related to it.
- What contacts (customers, employees, suppliers) it has.
- What locations (sites, warehouses, drop ship locations, digital locations) it has.
- What tasks (projects, todos) there are. I see 'service board implementation' which is an old project on 'Completed' status there right now, for example. And if I drill into that project, then I can in turn see every entity that was ever linked to it.
- What messages (emails, phone calls, etc). there are. Phone calls, SMS etc. are not integrated yet.
- Blueprints (e.g. 'Monthly vendor reconciliation 3rd day of every month'). auto rules not implemented yet, I'm manually clicking a 'load in template' button.
- Etc.
I can also view(/filter/etc.) by entity Status.
- 0. New
- 1. Respond (Something has changed, e.g. a customer has replied into it. Attention is needed).
- 2. Active <-- this position is not a hill I am prepared to die on. Personal preference.
- 3. Waiting (on external)
- 4. Hold (no action to take until specific date)
- 5. Evaluate (no further action needed apart from analysis+closure)
- 6. Cancelled
- 7. Complete
So everything actionable floats to the top, and everything non-actionable sinks to the bottom. When an entity becomes actionable, it then automatically floats to the top (e.g. 'EmployeeA was supposed to have provided reportX by now. Followup'.)
But basically, 2 clicks to create an entity. 3 moves to relate an entity to another entity. Automatic entity creation from different services into Supabase (Gmail, working on MSGraph, etc.), with server side triggers & functions to process and automate.
I can view the entity-relationship graph via 'displays' (not all implemented yet):
- Pods (drag and drop widget grid)
- Kanban
- Form
- Table
- Calendar
- Map
- Json
- CPA / Nodes
- etc.
So it's just a representation of my brain that saves a lot of time (especially in tracking things down) and augments my own abilities.
I am working on using this in a B2B product (an ERP for MSPs is the primary use case).
I consider every ticketing system, ITIL service board app, todo app, messaging system, Jira, PM application, ERP, etc... To be completely missing the big picture.
As soon as someone has implemented both a 'Customer' and a 'Supplier' table in a RDBMS, I think an opportunity has been missed and that we have not realised how refined relational db tech has become in the last 10 years.
I consider us to need a generalist 'Hub' structure that other services interface with. Self hosting option vital.
I have formal data structure training (relational database design, yes a lot of UML historically, set theory), consider this abstract level of the space to be slightly more objective than others do, and have the hubris to consider myself qualified. But, I know that I would not be able to convert everyone to my way of thinking - Nor do I have the time or the inclination. I just love thinking about data structures.
Preempting possible 'wtf's: Yup, relational db. GraphQL also fine. Relational db will not necessary avoid n+1 problem. Equally, n+1 problem is OK in GraphQL (I don't consider it a deal breaker and support the tech inc. in context here). KV db tech not appropriate.
Security & Encryption on row/property level = possible but significant challenge to keep performant (But I suggest that it is feasible, even in 5+TB tables). Limited stress testing so far has had surprisingly good results. Due to atomic entity-relationship granularity, it makes selective role-access very powerful. The problem then becomes the shape of the graph (the structure itself - not the data - of the graph, if exposed, could imply information and is a huge security risk). The other major risk is dynamic querying.
Data should be siloed to the extreme. Added benefit of making hosting changes hot-swappable. Go self hosted in (almost)singleclick. each 'space' needs own tables (own schemas not really relevant for security). Also added benefit for performance. 'sharing' is 'opt-in' (with caveat/nuance for automation), and explicitly 'crosses the border' to go outside of a space. This is at the cost of inter-space queries, but that's a necessary loss.
All current solutions on the market fail security sniff tests.
Very eager to get local LLM hooked up to my personal graph for natural language querying (I think my dataset+datastructure is the most appropriate I have ever seen, for this. AI totally outside of my wheelhouse though), but it won't bubble up to the top of priorities for a few months at least
This is super interesting, thanks for the detail! I bet the LLM experiment will be very interesting for you and I hope you share the results here on HN!
This is me. I eventually got to a place where I was also struggling to deal with _multiple_ todo lists all over the place. I felt like I was either in a place to just email myself todos or change up how I used my inbox.
I went with the latter, so now if an email requires action but not immediate, I add it as a task in omnifocus and link to the email. It’s not perfect, but it least it’s a step towards consolidating my todo list for email.
I did go with a label/folder todo approach initially, but I found myself just hating to go there.
> most?) of my job involves reading an E-mail where Alice has a problem and needs me to E-mail Bob to get them to do something.
Why are you the intermediary? Could you instead just connect Alice with the right person (in this case Bob) and they can work through the detail? Why play telephone?
LOL I ask myself this often. In a perfect world where everyone knows everyone and where everyone has the time and energy to hound everyone else, this part of Project Management wouldn't even exist. But in big companies, people get siloed and isolated from other teams, people get busy, they forget, so they need a juggler like me to keep all the balls in the air.
For your own sanity, don't be that in-between. It's fine to be the person they ask, but use that as an excuse to introduce the parties and then step out of the way.
It's good to use your domain knowledge to make the company run more smoothly. It's not so good to become the chokepoint, or the company directory.
There’s an interesting product idea (that’s maybe a dead end): make email inboxing more dependency-graphy. I do often find myself expecting an important email and having a known set of follow-on tasks
Yeah. Search isn't the issue. Even inbox zero people archive all their emails anyway. They don't delete them.
Inbox 20,000 makes sense for a lot of people btw. Depends on your work and how important being organised with your email is to you.
Many people just declare email bankruptcy and don't care.
If you are on a very high number of emails, you can just archive the last 19,800 and go from there.
And then unsubscribe from a lot of your key email offenders.
gmail has a snooze feature, so if you can snooze todo things if you know you can't make immediate progress on them if you don't mind a bit of delay in getting reminded to do things.
the web interface also has an add to tasks button if you use their latest todo list app, but i haven't found it all that useful
I'm a firm believer and practitioner of the 'inbox zero' method of email management. Naming this inbox helper/organizer the same may confuse people. Why did you choose that name? Seems a disservice to the method you're trying to promote.
In the defence of the creator, I assume they're goal was not to create a "newsletter manager", they probably just saw that as low hanging fruit to relieve some of the pain of managing an inbox.
I can easily see them adding more and more features which improve email and nudge users to inbox zero.
I think this is a decent starting point, and for a repo that is only 5 months old, it looks like they've done a decent job of understanding a good UX, and showing the user the emails they are most likely going to want to archive, etc.
The testimonials by "@james", "@yoni", "@ktyr" are pretty obviously not real just from their implausibly short Twitter handles. The actual Twitter handles have different profile pics that simply don't match the ones displayed on the landing page. For example @ktyr is not an Indian male but an Asian girl who hasn't tweeted since 2012.
@prem_saini1 also is not the same person.
@emarky is the same person, but he doesn't seem to have tweeted the testimonial.
As I said, I'm just really bored :)