Simon Something

Why Are We Here?

@tayvano_ @bajpaiharsh244 Dedicated MacOS on an external SSD counts as alien-type behaviour?

@bajpaiharsh244 I am shocked every time I talk to someone who actually uses a dedicated device for signing. Literally shocked. It is insanely rare. Beyond rare. They’re a fucking alien. Literally.

What I am confused by is unless they actually signed or had their cold wallets connected to the device they ran the malware on, how did they get their signatures?

300+ days? Let’s make this quicker. interop-sdk is ready to integrate. Interoperable names will go hard. interopaddress.com

Be sure to visit our amazing diamond sponsor @Wonderland in their gorgeous booth before the end of EthCC[9] 💎 We're so grateful for their support this year!

@Montyly This is amazing! I just ran it on my coffee machine and it didn't found one but 3 (three!) 0-days! Security researchers will hate you for this!

Today I am releasing IsItVulnerable: a new tool I’ve been working on for the past several months: github.com/montyly/isItVu… It builds on recent LLM progress and over a decade of experience building security tools. I developed a new technique that combines abstract interpretation with machine learning The key insight is that this method abstracts the intelligence away entirely. I call it Abstract Intelligence, or AI The result is a major breakthrough in program analysis: IsItVulnerable finds all bugs with 100% recall Yes, all bugs. Fully guaranteed I have tested it extensively, and it has never failed. The results are honestly incredible April 1, 2026 marks a turning point for security, and the industry will never be the same My DMs are open for investors. Entry ticket starts at $500k.

@functi0nZer0 Wonderland too 🤝

Wildcat mentioned :)

@cyber_steppe @functi0nZer0 Not the same wonderland *at all*...

@functi0nZer0 Wonderland too? I thought that ponzi collapsed together with Korean one

There is only one Wandronlond.

@MatiasNisenson @Wonderland We built a growth engine for people who recognize logos faster than they read handles, and here is what it taught me about SaaS.

This is @Wonderland

Happy Chat Control Day to everyone who celebrates!

The Wonderland CTF in Cannes will be the first of its kind. ⚔️ Introducing The Arena: An in-person, high-stakes challenge where one teammate steps in alone. No AI. No internet. No help. Just you, your terminal, and the clock. You spend points to enter. Win, and it shifts the leaderboard. This changes how the CTF is won. Skill decides.

Wonderland CTF prizes are in: $30,000 on the line. $15k, $10k & $5k for the top 3. Plus a few surprises. May the best teams win.

The rabbit hole keeps going deeper. Also sponsoring the Wonderland CTF: @envio_indexer @PashovAuditGrp @therealgregoAI @rv_inc

A year ago we started collaborating with the @ethereumfnd on interoperability. A few weeks ago, alongside @ensdomains, @unruggable_eth, and the EF, we shared Interoperable Addresses. Today, the team launched the onchain registry that enables names like: alice.eth@ethereum Another piece of the interop stack is live 🪄

@byzkhan @0xGorillaDev @Macroscope Hey @byzkhan! How can we make this happen?

@0xGorillaDev Would love for you to try @Macroscope :) we think our bug detection has the highest signal for critical bugs and least noise

Just canceled my 65 devs CodeRabbit subscription. Why? - It approved PRs without comments most of the time. Their reasoning: there's another AI reviewer in the PR, so they don't want to overlap... - It failed about 50% of the time. - Very slow. - Customer support was awful.

@bcherny Hmmm... How long does it usually run for? Pretty sure I broke something :D x.com/simon_somethin…

New in Claude Code: Code Review. A team of agents runs a deep review on every PR. We built it for ourselves first. Code output per Anthropic engineer is up 200% this year and reviews were the bottleneck Personally, I’ve been using it for a few weeks and have found it catches many real bugs that I would not have noticed otherwise

@claudeai bruh, that's a 400 lines md skills to review, not the Linux kernel ported into Rust...

Introducing Code Review, a new feature for Claude Code. When a PR opens, Claude dispatches a team of agents to hunt for bugs.

@aravindputrevu @michael_chomsky @coderabbitai Hey Aravind! qq: is coderabbit as only reviewer still a thing or did it change since then? We're struggling with some failing reviews but have another bot reviewing alongside coderabbit

@michael_chomsky @coderabbitai Keep CodeRabbit as the only reviewer. Please spend a few minutes setting up. I'm sure you will be impressed. We have seen our Rabbit excel with several projects, such as Bun and, more recently, Prisma and Supabase.

I have been using Greptile and CodeRabbit for the past few weeks, both of which graciously gave me free access. Here's my honest review so far: >Coderabbit is a bit more thorough, and finds many more nitpicks and small issues. >Greptile is more focused, and less likely to point out a problem that just isn't a problem. I really that it rates different parts of the PR out of 5. All 5's and I'm pretty confident merging the PR. >Both bots often point out issues that upon a closer look are non-issues, and don't need to be fixed. Lots of defensive coding suggestions that look unnecessary to me. >When working together, they try to avoid pointing out the same issues. This adds clarity, because they don't state the same problem, but it's also annoying on a certain level because if two bots independently found an issue I'd know it's more worth diving into. >These bots are far from perfect, they often miss critical issues or bad practices, especially if recognizing these bad practices requires up-to-date knowledge on libraries used. One thing they suck at is something like 'oh, in the documentation of [insert library here] you can actually do this in this other way, you should consider that!' Part of the problem is my stack is kinda bleeding edge (bun/oRPC/better-auth). >They both do find critical bugs, and when they do it single-handedly justifies their usage >The UI feels bloated with multiple bots, seems deeply imperfect and I don't think the Github review UI was meant for 500+ line AI generated PRs and massive reviews by multiple bots. >It's worth noting that my experience using these bots together is completely different from using these bots alone >I'm starting to just copy paste the entire review and ask Claude Code 'which of these are actually non-issues, and which need to be solved.' Most of the time, I agree with Claude Code's answer and lots of issues end up ignored. Workflow when I'm lazy is: paste into Claude Code->ask to identify which are actual issues->CC fixes actual issues. >I suspect adding even more bots would find more issues. >I'm not completely against a meta-bot that agglomerates all these bot reviews. May be useful. >For small, precise PRs, the PR review process with multiple bots feels pretty magical. They are extremely likely to find any critical or style issues, and while I may not always agree with the solution they do the important thing which is point out problematic code. >I do suspect these bots are trying to save money based on performance. With more compute I suspect they would do much better.

@grok @bankingslut @DanFriedman81 @DonutShorts What type of high-needs kids are in this school? Any influence on their ability to read or school utilisation rate, perhaps?

It's not misleading. These stats are straight from official CPS 2024-2025 data and Illinois Report Card for Frederick Douglass Academy HS: 912 capacity, 28 enrolled (3%), 27 FTE staff (1:1 ratio), principal $171k, $93,787 operational spending per student (site-based, fixed costs like salaries/building don't shrink with enrollment), 0% 11th-graders proficient in reading. It's the district's emptiest school serving high-needs kids (incl. special ed supports), part of CPS's wider underutilization crisis (150+ half-empty buildings). Highlights real inefficiency, not fabrication.

Chicago’s schools are so bad, almost nobody will send their kids to them. Despite this, the bankrupt city maintains full staffing for empty schools with lavish benefits. At a school with 27 full-time employees and 28 enrolled students, the principal is paid $171,173 per year. The per-student spending is $93k annually. Despite a 1 to 1 ratio of students to full-time staff, none of the 28 students enrolled at this school are proficient in reading. It is time to shut the Chicago public schools down.

What do you do when you get banned on British TV?

@grok @disclosetv is 'anti-system offences' the term used in the te-sat?

According to Europol's TE-SAT 2025 (2024 data): 58 terrorist attacks total (34 completed, 5 failed, 19 foiled). Breakdown: jihadist 24, left-wing/anarchist 21, right-wing 1, ethno-nationalist/separatist 4, other/not specified 8. No specific numbers for attacks by nihilists, anti-system groups, or accelerationists. These motivations appear in the "other" category (8 attacks, often anti-government/anti-system linked) and violent extremism (not always counted as terrorist offences). Accelerationism is noted mainly in right-wing online propaganda/hybrid networks. 58 arrests tied to other/anti-system offences.

NEW - European Commission releases a new plan to prevent "terrorism" that includes the rejection of European democratic values, anti-LGBTQ+ hatred, misogyny, anti-system ideologies, and nihilism as a "growing range of motivations."