Hey hey 👋 Lots of new people recently, so let's get to know each other
I'm Stu. I'm a principal engineer who has spent years at places like Checkout and ClearBank. Whilst I'm building my future, I share interesting things and what I learn on the way
What are you working on?
Silently dropping from 20x tier to 10x tier is not great at all. It's almost getting to a point where an open harness and a fleet of open weight models because a better option...
We’re renaming our Max plans:
Max 5x → Max
Max 20x → Ultra
Max is still $100/month, and Ultra is still $200. If you’re a subscriber, there’s nothing you need to do.
We’re also removing the 5-hour limit for verified Ultra accounts, starting today with most existing subscribers. Newer accounts keep a temporary limit until they’re verified, which helps us prevent fraud and abuse. You'll get an email when it's removed.
Weekly limits remain unchanged. View our plans here: https://t.co/niRpyGF5Me
Small one if you're building a sign-in code screen. Use one input with autocomplete="one-time-code" and style it as six boxes, rather than six separate inputs
Six inputs looks the same but paste and autofill usually end up half broken. Just did it this way on the new project's sign-in pages and it's so much nicer :)
TIL about Datadog's pathfinding project. It documents pretty much every known AWS IAM privilege escalation path, what permissions each one needs and how to spot it
Honestly worth a look if you touch AWS at all
@bas_fijneman Nice nice!! Anything we can share a sneak peak of?
i’m just cooking away, starting a new project so i can port OneHaus across to it. So just in that exciting early stage part :) will launch the homepage and start posting later this week hopefully
Connecting to an MCP server can be enough. Two MCP Inspector advisories, Oct 21:
- Server-chosen OAuth URL runs as PowerShell on Windows/WSL (high; fix 2.12.0)
- OAuth callback flips transport to stdio, runs a local command (critical; fix 1.0.3)
Do you vet MCP servers first?
TIL putting links in a reply instead of the main post isn't really a thing anymore. X said in July you don't need to bother
Updated @mcpscores now to see if this holds true
@tarasshyn Also, not everyone can create software that scales. handling a few hundred requests/sec is trivial, going to thousands and beyond becomes somewhat more complex. In some aspects could != should
@GeoffreyHuntley@bot dr eggbot is great. Followed your recommmendation as i started fresh with grok bot - it’s really good. Just told it what i wanted to do and various things got hooked up. First time a bot felt good to use
The "best household management apps 2026" post is still the second most visited page on onehaus, behind only the homepage
People starting the free cleaning schedule tool went up about 65% this week too. Slowly starting to get why everyone says just keep writing content :)
Finally merged my staging to main PR for onehaus. It'd been open since 26 Sep and closed 278 issues 🙈
Note to self, maybe don't leave it a month next time
Building verifymcp has been way harder than I expected. Every MCP server gets scanned from a clean slate, so that's a full package restore every single time, across thousands of servers, on a pretty small box
CPU sat at 60-70% for days. A few changes later (caching, fewer retries, only rescanning when a package actually changes) and I think it's finally under control 😅
I honestly can't remember another week quite like October 14–21 in software security.
Just between October 14 and October 21:
- Next.js patched two critical RCE vulnerabilities (9.5/10).
- NGINX patched five security vulnerabilities.
- PHP disclosed critical RCE vulnerabilities. AWS observed exploitation attempts four days before disclosure.
- Node.js patched multiple vulnerabilities, including potential RCE and TLS validation bypasses.
- GitLab disclosed another 10/10 vulnerability allowing potential unauthenticated RCE.
- LiteLLM's compromised releases distributed credential-stealing malware.
- Railway disclosed a potential exposure of customer secrets and API keys.
- OpenCode reported a breach affecting 1.4% of its Console users.
- eMudhra had eight trivially recoverable certificate authority private keys. One of them was literally the number 4!
Also, Microsoft reported AI-orchestrated attacks targeting over 2,250 Microsoft 365 tenants, with attackers obtaining privileged access in under three hours in multiple cases.
CrowdStrike documented AI agents conducting real-world intrusions, including one that obtained domain-controller credentials in approximately 20 minutes without further human guidance.
Meanwhile, Cloudflare observed PHP exploitation attempts growing to millions within hours of disclosure. Some of the attacks even started before any public disclosure, which shows that malicious actors have ways to get to the important information way before any organization does.
All of this obviously made me think about what this means for engineering organizations.
For years, we've treated vulnerability management as something that happens periodically. We review advisories, create tickets, plan upgrades, and eventually deploy patches. Sometimes days later, sometimes weeks, and quite often months or even years.
That's not going to work anymore and technical decision-makers will need to fundamentally rethink how their organizations approach security.
Continuous monitoring, immediate exposure assessment, automated mitigation, credential rotation, and rapid patching need to become standard engineering capabilities.
We need to start measuring our security response times in minutes, not days, weeks, months, or years.
Last night I got Grok Bot to sort through my emails and it cleared over 11k and organised the rest for me. Honestly impressed
Far easier to set up than it was with Hermes