This is really cool - I think desktop tools, like native apps, CLIs, self-hosted tools, etc., have a lot of potential for innovation since they can be consumed by agents and the user can maintain more privacy, while also using their preferred agents and not being locked into a specific one.
HN user
RichardChu
Building https://fluxmail.ai
What's the use case for doing that, just wondering?
Awesome! Let me know if you have any feedback.
Thanks! I'm planning on supporting a granular permission system, so per account you should be able to customize which tools it should be able to use (e.g. readonly, read+write including drafts (but no send), or full access).
You should be able to customize the permissions per API key as well, so if there's a case where you want to write or send something just one time then you can just swap in the API key and do that and swap back if you want to.
There will also be a dashboard (in addition to the CLI) so that you can easily audit everything that has happened in the system.
I'm working on a self-hosted MCP server to give your AI agents access to email: https://github.com/churichard/fluxmail-mcp
The main reason is because I wanted a privacy-preserving way to access my email without using ChatGPT or Claude or another hosted email connector. Also, this supports connecting multiple accounts, even multiple accounts from different email providers, with a unified API. And you can use it with any AI agent, even one you build yourself.
It supports Gmail API and IMAP/SMTP right now, with Outlook / Microsoft Graph coming soon.
Better Auth is great, I use it for all my projects. Congrats to the team!
Is this AI written (or edited)? The word "genuinely" appears 4 times on the page.
Man, I hate how often people/LLMs use that word now. Maybe other people gloss over it but it's super distracting to me.
I have qualms with Spotify as well, but not sure I entirely agree with the premise of the article. It still takes effort to find good music. There's a LOT of music being produced nowadays, and wading through all of that and finding niche stuff you like can still be a "hunt" - it's just a different kind.
I'm working on Fluxmail, an AI-native email app for people who get a lot of email. It has an AI agent, automatic prioritization and categorization, automatic follow-up reminders, drafting with AI, and a bunch of other stuff.
There's a couple of other ideas I'm integrating right now, like a kanban view for your emails
I'm working on an AI-native email client that organizes, prioritizes, and drafts emails for you.
The vision is for everyone to have an executive assistant that manages their email. It's built for people who spend hours in their inbox every week.
It has automatic prioritization, split inboxes, snippets, bundles, automatic follow-up reminders, and an AI agent that can do stuff for you -- without deleting your emails.
If you've read this far, I'd encourage you to give it a try and let me know what you think!
I'm working on Fluxmail, an AI-powered email client! https://fluxmail.ai
I'm building a new AI-powered email client called Fluxmail: https://fluxmail.ai
Here is the current list of features:
- Unified inbox: Connect multiple Gmail accounts and see all your emails in one inbox. No more switching between different tabs or accounts.
- Email bundles: Automatically group your emails together into bundles, so if you have many emails from the same sender, they'll no longer clutter up your inbox.
- Email summarization: Save time by getting summaries of your email threads.
- Keyboard shortcuts: Navigate through your inbox using Gmail-style keyboard shortcuts.
- Automatic dark mode: All your emails are converted to dark mode automatically.
- Block email trackers: We detect and block email trackers to protect your privacy.
- Privacy-first: We sync your emails from Gmail and cache them locally on your device - they are not stored on our servers. We don't train on your data or use AI providers that train on your data.
Lots of other features coming soon, like split inbox, AI search, etc. It's in early access now, would love feedback from anyone interested in checking it out!
I'm also curious to hear - what is the most annoying thing about email that you have to deal with today?
It should be fixed now. Thanks for flagging!
I'm building a new email client called Fluxmail: https://fluxmail.ai
You can connect a bunch of Gmail accounts with it and see all your emails together in one place. It also has some features to better organize your emails and automatically categorize them, as well as some AI features for helping you to draft emails.
It's in early access now, would love feedback from anyone interested in checking it out!
It's an AI-native email client. Launching soon!
My goal is to help people get done with email faster, so that they can get back to doing other stuff. A lot of the features are designed around this goal: unified inbox, AI summarization, AI email drafting, etc.
Some of these are table stakes but I think there's also an opportunity to significantly revamp how email is done in the AI age. Imagine having your own personal assistant that goes through your email and surfaces the highest priority things that you need to know automatically.
Primarily bundle size. Pglite is a 3 MB binary, SQLite is <1 MB.
I was evaluating local storage solutions a while back and I tried setting up PGlite, but unfortunately I couldn't get it to work in Next.js with Turbopack in a web worker.
I've been using SQLite locally instead with wa-sqlite and it's been working great for my use case so far. It's also more lightweight.
I'm working on https://www.fluxmail.ai, a modern, AI-powered email client.
I think there's a lot of potential for AI to improve the way we organize and manage our inboxes, while still letting us keep control over it.
What I've learned is that there are a lot of little features that make up a good email client that you may not even think about when using one, like threading, quote blocks, even what email address(es) to autofill when you reply to an email. For an app you use potentially for hours a day, the polish and "last 20%" makes a huge difference - and takes a while to build!
If you have any feedback, especially on what features are most important to you in an email app, I'd love to hear it :)
I'm working on 2 projects right now:
1. Fluxmail - https://fluxmail.ai
Fluxmail is an AI-powered email app that helps you get done with email faster. There are a couple of core tenets/features that it has, including:
- local-first - we don't store your emails and we make interactions as fast as possible
- unified inbox - so you can view emails from all your email addresses in one place
- AI-native - helping you draft emails, search for emails, and read through your emails faster
I'd love to hear if these features resonate with you, or if there are other features that you feel are missing from your current email app.
2. ExploreJobs.ai - https://explorejobs.ai
This is a job board for AI jobs and companies. The job market in AI is pretty hot right now, and there are a lot of cool AI companies out there. I'm hoping to connect job seekers with fast-growing AI companies.
I'm working on 2 projects right now:
1. Fluxmail - https://fluxmail.ai
Fluxmail is an AI-powered email app that helps you get done with email faster. I think there's a significant opportunity for AI to change the way we use email, and I'm experimenting with ways to improve the status quo. I'd love to hear what features you'd like to see in such an app!
2. ExploreJobs.ai - https://explorejobs.ai
This is a job board for AI jobs and companies. The job market in AI is pretty hot right now, and there are a lot of cool AI companies out there. I'm hoping to connect job seekers with fast-growing AI companies.
https://notabase.io - a note-taking app for networked thinking.
It supports page stacking, linked references, block references, a graph view, and all that good stuff. Think of it as similar to Roam Research / Obsidian.
It's also open source so you can self-host it. Here's the code: https://github.com/churichard/notabase
I'm hoping to add support for shareable links soon. Open to other ideas or feedback!
When evaluating rich text editors for the note-taking app I started about a year ago (https://notabase.io), I ended up going with Slate because of its flexible schema and customizable plugin architecture.
I sort of regret that choice now. I ran into a lot of bugs when integrating it which I had to manually work around; issues go months without being addressed; and there still isn't good cross-platform support, especially for Android. With a more active contributor base, Slate could be a fantastic library, but I get the feeling that it's in maintenance mode now, with not many major changes in the past year and a v1.0 still far in the future.
Tiptap looks like it might be a good choice now, but I find it off-putting that I can't insert links in the demo editor on Tiptap's website (https://tiptap.dev), especially for my use case (a note-taking app whose core concept revolves around links).
I'm working on an open source note-taking app called Notabase [1].
It's built primarily for my use - I just never liked most existing note-taking apps and wanted to make one that's easy to use and fit the way that I think. I made it open source [2] so other people can build on top of my ideas, and released a hosted version so that other people can use it if they like it.
It would be nice if other people found it helpful, but regardless it's something that I intrinsically enjoy working on.
Maybe it's unclear, but you just need to insert a row in the subscriptions table with the value "ACTIVE". You don't actually need to have a real subscription on the hosted version of Notabase.
Besides, if you know how to code, it's simple to change the max number of notes to something like Number.POSITIVE_INFINITY anyway - but adding a row in the database is likely easier so that's what I recommend.
It may be a bit of a hassle, but I do this because the codebase for the hosted version and the self-hosted version is the same, which makes it easier for me to maintain. In the future, I may make a separate self-hosted version so that this step can be skipped.
Yep, although there are people who would change the license or ignore it, and hope that no one notices.
Hey, the creator of Notabase [1] here, which it looks like MdSilo was based off of.
Your project looks cool! Always interesting to see how other people tackle note-taking. Thanks for keeping it open source.
I'd encourage people to check out Notabase as well, which offers an alternative vision and UI which some people might like better. You can use it hosted, or self-host it yourself -- the code is open source [2].
Best of luck with MdSilo! :)
If you want an open source alternative, you should check out Notabase [0]. It's like a mix of Notion + Roam Research, designed to be delightful and easy to use. The difference from Obsidian is that it's database-backed.
Might be worth checking out Notabase [0]. The UI is designed to be really clean and easy-to-use.
I'm making Notabase [1], which is a personal knowledge base for non-linear thinking. It's inspired by note-taking apps like Notion, Roam Research, and Obsidian.
I started it because I wanted a note-taking app for networked thought / Zettelkasten that is easy to use for the average user (not the average Hacker News reader). So you don't need to use Markdown if you don't want to, you can format text in other ways like using the formatting toolbar or through keyboard shortcuts. And syncing between different devices is handled automatically for you since it's a web app.
I also wanted to make it open source [2] so that anyone could look at the code and contribute to it if they wanted to. I'm not a big fan of the way that a lot of the apps in this space are closed source -- you don't really know what they're doing with your data, and it doesn't inspire confidence in the app's longevity.
It's still early in development, but I'm hoping to build a solid community behind it.