I noticed that too. It's unreasonable to extrapolate anything from that graph when it only ends at 2022.

It might be a joke, but I'm sure there are lots of people who don't perceive it as such. For me, it pulled me out of the article and diluted its advice.

The funny thing is: this website reads like AI.

I'm working on self-hosted local email infrastructure for agents and apps: https://github.com/churichard/fluxmail

It consists of a MCP server, a CLI, and a REST API. Everything is run and hosted locally on your computer. You can use the MCP server to give your AI agents access to email, use the CLI to interact with your email, or build your own email apps on top of the REST API.

The goal is to make it simple and easy to interact with your email through any interface you choose, and to give you control over the way that you use your email.

It supports Gmail, Outlook, and IMAP/SMTP. Give it a try and let me know what you think!

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.

very cool, I dont think email is going anywhere. im going to give it a try. im working on a tool for myself that scores emails with custom labels, auto-unsubscribes, report+block+filter anyone that doesn't respect unsubscribe

Awesome! Let me know if you have any feedback.

Really great idea. What are some of the access controls available

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.

for that sort a tool it would be nice to restrict access to specific labels , or API query patterns.

What's the use case for doing that, just wondering?

Reducing scope . For example , preventing private emails from being accessed by the model