kill -9 2025

You shouldn’t have to ask an agent to help.

Like any good employee, it should solve the problem before you know there is one in the first place.

Proactive agent sees this message, automatically drafts a ticket, and then sends you a quick message:

"I just drafted a ticket for you based on your conversation. Want me to create it"?

All you have to do is say yes.

Imagine you're in slack and your boss starts chatting about a bug they noticed.

They say, "Can you create a ticket?".

Proactive agents listen.

They comb through all your operational data, extract intent, draft work, request approval, and then finish the job.

Instead of having to ask AI what you want it to do all the time, now you only need to review and approve drafted work.

"Hey, I noticed you have 3 unread emails that look important, can I give you a summary and then review some replies I drafted for you?"

AI becomes the initiator, just like "Her".

With Proactive Agents.

First, give AI access to a firehose of data:

- Slack threads
- Github commits
- Jira Tickets
- Emails
- Calendar events
- Meeting transcripts

Now our AI agents can connect to apps and control them, but we're still prompting the AI, repeating ourselves over and over.

So how do we kill prompting once and for all?

We needed to move from agents as information-gatherers to agents as action-takers.

So we created tool calling. MCP captivated us all.

So we connected LLMs to external data sources and gave them reasoning capabilities.

This unlocked deeper research, with accurate answers and timely information.

But how is that research useful if you aren't able to *do* anything with it?

AI started isolated and offline. You asked, it answered based on it's internal world knowledge.

But this often led to outdated or inaccurate answers due to knowledge cutoff

What if you never had to prompt AI again?

Proactive agents will be the the first time we truly feel the AGI

Let me explain 🧵

easiest way to mcp is with skeet.build

mcp to update the team in slack about your progress on a ticket

I happily pay for TablePlus, Medis, and Toggl.

I was using the free tier enough that the paywalls got annoying, so I converted :)

Definitely worth it!

Rectangle Move and resize windows in macOS using keyboard shortcuts or snap areas. The official page for Rectangle.

Window management is key, rectangle is free and makes it easy

Most common shortcuts:
- Swap a window to another monitor
- Make a window full screen
- Move a window to left half or right half of screen

rectangleapp.com

Toggl Track: Time Tracking Software for Any Workflow The world's best time tracking software. Track employee time, send beautiful client reports, and calculate profitability. Free and paid plans to save you time.

I track my time meticulously to see where I'm spending it (and for consulting work too)

Toggl is great because it's really easy to turn timers on and off. There's a great raycast integration, and live activities on iOS too.

toggl.com

Makes viewing and debugging your data in redis really easy. Works great for local data and hosted redis in prod too.

getmedis.com

TablePlus | Modern, Native Tool for Database Management. Modern, native client with intuitive GUI tools to create, access, query & edit multiple relational databases: MySQL, PostgreSQL, SQLite, Microsoft SQL Server, Amazon Redshift, MariaDB, CockroachDB, Vertica, Cassandra, and Redis.

TablePlus is great for running SQL snippets and inspecting staging / production data quickly.

I prefer it over looking via supabase or querying directly

tableplus.com

Some software I use every day but I rarely see mentioned:

TablePlus - GUI for your database
Medis - GUI for redis
Toggl - Time tracking (works great with Raycast)
Rectangle - Window management for MacOS

Links below 👇

cursor is insanely powerful now with all these tools added

Skeet - Connect Apps to Cursor Skeet - One Shot Coding Workflows

The possibilities are endless.

MCP lays the foundation for extremely powerful agentic workflows, where all your software and data is accessible.

Experience this new future today with skeet.build

Suddenly ALL of the software you use is interconnected.

Apps talking to and controlling other apps.

- Google flights can check your calendar
- Cursor can fix production bugs from Sentry
- Customer feature requests are prioritized in Linear and sent to Slack

MCP is about as cool as HTTP, so who cares?

You don't have to understand HTTP to use one of the greatest inventions ever: the internet

MCP is the same. It's not about the protocol, it's about what the protocol enables.

Skeet - Connect Apps to Cursor Skeet - One Shot Coding Workflows

This is why we’re building skeet.build—so AI can integrate seamlessly with the software that runs the world.

The exciting thing about this shift is it means it is more important than ever to focus on user value.

If you're a builder, as long as your focused on solving burning customer problems, you will continue to win.

The solutions will definitely look different than they used to.

A user interface is a means to an end, doesn’t matter if it’s an app UI you built, an external LLM, an ai agent, or even a brain computer interface

Protocols like MCP make this possible. The earlier your app integrates into the AI ecosystem, the more leverage you’ll have.

Some apps won’t even need a frontend 🤯

They’ll just structure data, run workflows, and expose LLM-friendly APIs.

Just because you aren’t using a steering wheel in a self driving car doesn’t mean it isn’t getting you somewhere

But APIs change, integrations break, and the UX of unified apps never lived up to the promise.

Now, LLMs can just talk to your apps for you. Instead of every app needing its own UI, an LLM becomes the interface—understanding natural language and controlling tools directly.

AI will drastically change UI/UX design

Natural language is one of the easiest ways to interact with a computer.
Most people don’t know how to code, but everyone knows how to talk.

Apps used to compete on integrations— trying to make hundreds of workplace tools all cooperate.