I like making the next step explicit while the first agent is still working. A separate review chat gets the diff to critique, instead of a vague “done” message to trust.
The useful bit here is coming back to a clear question instead of a mystery half-finished run. Answer it, and the agent continues where it paused.
No repo setup for invoice totals, supplier quotes or top sellers. I like this as a practical Cowork test: three separate jobs running side by side, with something concrete to review at the end.
I’d rather give the agent the Figma file than keep sending screenshots back and forth. This is one of the more practical uses of MCP.
Mike retweeted
A tour of Unstoppable Code, top to bottom.
🔍 Search across the whole workspace
🧭 App Advisor, ask how to do something and it does it
⏰ Automations that wait for an event and ping you
🔌 Plugins and MCP, bring the tools you already use
⚙️ Pipelines, build a workflow once and run it forever
🎛 Set the app up the way you work
🌳 Parallel coding agents, each in its own worktree
💬 Cowork for the work that has no repo
We are in the midst of the Claude Code moment for browser use.
Agents are finally good enough to reliably complete arbitrary and medium-length knowledge work tasks in a browser, and this is what's behind the "aha" moment for grok bot
Cloud agents infrastructure has also gotten a lot better, which paves over a lot of the issues of running scheduled tasks locally
Just today while prepping for my burning man trip I had a lot of difficulty making sure my socials and outbound automations would continue to run while I'm offline. Things like:
> If an automation runs in the cloud I don't have to worry about being offline
> Sharing the automation with a teammate required dumping him an md file with memory and specs for him to ask his instance to setup locally. This will soon be 1-click to share for @unstoppableweb users
> Importing your chrome passwords into the agent's browser so you don't need to enter them manually (coming soon)
> Upgrading our browser use to use WebMCP, a growing standard for easing browser use
Things are moving fast! White collar workers are just now having their reckoning with AI. We aim to bring this new tech to non-technical users in an easy-to-digest way so they can start their self improvement loops quickly
my own workflows are changing by the day and I'm loving all the new leverage I'm getting.
I also predict social platforms will fight back here as the tech is distributed more broadly
Exciting times!
I'm back in today after a week off. On my last working day, I linked my Slack, Gmail and Linear to Unstoppable Code.
My process to catch up on the previous week was asking it: "Can you look through Slack, Gmail and Linear for the past week and tell me what I need to know/work on?".
I would have spent half of today catching up and trying to pull out what I needed to know without this.
I don't really want a dev tool that needs me to dig through docs every time I need to use a new function.
The App Advisor in Unstoppable Code is quietly useful for that.
Ask where to set up a watcher, how a pipeline works, or what a setting does, and it answers inside the app.
I've spent this week moving more of my workflows over to Unstoppable Code.
Before, I was using Claude CLI for Remotion videos, ElevenLabs via browser for voiceovers, Claude desktop with skills for script writing and Claude CLI to work on a website.
Now, I've moved all of these to Unstoppable Code. I'm working in one window and running tasks simultaneously in their own workspaces.
Mike retweeted
Introducing Pipelines in Unstoppable Code.
For the repeatable work your team runs, make it once and run it over and over.
⚙️ Bundled presets out of the box, or build your own
🗣️ Describe what you want in plain language, the Pipeline Assistant writes the pipeline
⏰ Schedule them to run automatically. Set the cadence in plain language
🔒 Sensitive values stay in environment variables
🔁 Runs are durable and resumable
One of the strongest ships out of the box: a multi-agent code review that reviews your code across multiple providers and hands back a report.
👇 Try it 👇
Mike retweeted
Introducing Unstoppable Code!
An agentic platform that runs AI coding agents in parallel, each in its own isolated worktree, and Cowork tools for everything else. Free to download and free to start using.
🧵⬇️