惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

大猫的无限游戏
大猫的无限游戏
Webroot Blog
Webroot Blog
cs.AI updates on arXiv.org
cs.AI updates on arXiv.org
T
Threat Research - Cisco Blogs
V2EX - 技术
V2EX - 技术
L
LINUX DO - 热门话题
Google DeepMind News
Google DeepMind News
Recorded Future
Recorded Future
S
Schneier on Security
I
InfoQ
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
cs.CV updates on arXiv.org
cs.CV updates on arXiv.org
The GitHub Blog
The GitHub Blog
S
Security @ Cisco Blogs
O
OpenAI News
W
WeLiveSecurity
Vercel News
Vercel News
阮一峰的网络日志
阮一峰的网络日志
Simon Willison's Weblog
Simon Willison's Weblog
人人都是产品经理
人人都是产品经理
Cloudbric
Cloudbric
The Last Watchdog
The Last Watchdog
The Hacker News
The Hacker News
Google Online Security Blog
Google Online Security Blog
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
GbyAI
GbyAI
NISL@THU
NISL@THU
T
Tailwind CSS Blog
V
Visual Studio Blog
PCI Perspectives
PCI Perspectives
K
KPMG report finds enterprise disconnect between AI and its ROI | CIO
Jina AI
Jina AI
D
DataBreaches.Net
B
Blog RSS Feed
N
News and Events Feed by Topic
N
News and Events Feed by Topic
H
Heimdal Security Blog
cs.CL updates on arXiv.org
cs.CL updates on arXiv.org
腾讯CDC
Latest news
Latest news
V
Vulnerabilities – Threatpost
Hacker News: Ask HN
Hacker News: Ask HN
WordPress大学
WordPress大学
V
V2EX
aimingoo的专栏
aimingoo的专栏
博客园 - 司徒正美
Apple Machine Learning Research
Apple Machine Learning Research
D
Darknet – Hacking Tools, Hacker News & Cyber Security
The Register - Security
The Register - Security
Help Net Security
Help Net Security

Oskar Wickström

Coding on Paper Catching Typos on My Website with Browser Testing The Bombadil Terminal Experiment Computer Says No: Error Reporting for LTL A Year with the Daylight Computer Finding Bugs in a Coding Agent with Lightweight DST Machine: Learning; Human: Unlearning; How I Built “The Monospace Web” A Flexible Minimalist Neovim for 2024 Statically Typed Functional Programming with Python 3.12 Specifying State Machines with Temporal Logic Clearing Weeds and Planting Trees Introducing Quickstrom: High-confidence browser testing The TodoMVC Showdown: Testing with WebCheck Time Travelling and Fixing Bugs with Property-Based Testing Property-Based Testing in a Screencast Editor, Case Study 3: Integration Testing Property-Based Testing in a Screencast Editor, Case Study 2: Video Scene Classification Property-Based Testing in a Screencast Editor, Case Study 1: Timeline Flattening Property-Based Testing in a Screencast Editor: Introduction Why I’m No Longer Taking Donations Writing a Screencast Video Editor in Haskell Declarative GTK+ Programming with Haskell Finite-State Machines, Part 2: Explicit Typed State Transitions Modeling with Haskell Data Types Motor: Finite-State Machines in Haskell Automating the Build of your Technical Presentation Tagless Final Encoding of a Test Language Hyper: Elegant Weapons for a More Civilized Page Taking a Step Back from Oden Custom Formatting in HTML and LaTeX Code Listings using Pandoc Paramount Color Scheme for Vim
There and Back Again: From Quickstrom to Bombadil
Oskar Wickström · 2026-01-28 · via Oskar Wickström

Today I’m announcing and open-sourcing the Bombadil project — a brand new property-based browser testing framework. I started working on this when joining Antithesis two months ago. While still in its infancy, we’ve decided to build it in the open and share our progress from the start.

We decided on the name Bombadil last week. A few days later, this exploded:

It's going to be tough for startups when all the Lord of the Rings names are taken and the only thing left is something like Bombadil AI.

— Patrick Collison (@patrickc) January 25, 2026

While Quickstrom proved its worth, finding bugs in more than half of the TodoMVC apps, Bombadil aims to improve on its shortcomings while also envisioning a more ambitious future for generative testing of web apps: faster and smarter state space exploration, a modern and usable specification language, better tools for reproducing and debugging test failures, and a better distribution story.

I consider Bombadil the successor of Quickstrom. After years of trying to sustain Quickstrom through various models, I now have a much better answer: building it at Antithesis, making something valuable that is open and free to use, while also strengthening the company’s commercial offering. Bombadil can be used locally or in CI to test your web apps early. Power users can take it further and run Bombadil within Antithesis and its deterministic hypervisor. That gives you perfect reproductions of failed tests. You can even combine Bombadil with other workloads in Antithesis and test your entire stack deterministically. Isn’t that the holy grail of testing?

Bombadil is built from scratch with a focus on accessibility: a new specification language and better tooling for writing and maintaining specs. Right now, I’m working on a specification DSL in TypeScript. It’s based on linear temporal logic, just as Quickstrom, but aims to be a lot more ergonomic. Here’s an example that verifies that error messages eventually disappear:

const errorMessage = extract(
  (state) =>
    state.document.body.querySelector(".error")
        ?.textContent 
        ?? null
);

export const errorEventuallyDisappears = always(
  condition(() => errorMessage !== null).implies(
    eventually(() => errorMessage === null)
        .within(5, "seconds")
  )
);

Both the original hacky PureScript DSL and the bespoke language Specstrom were huge obstacles to adoption. Today, TypeScript is a widely adopted language among quality-minded web developers, and if you’re not into that, Bombadil will work with plain JavaScript too.

We’re writing Bombadil in Rust, leveraging its excellent ecosystem to ship a single statically-linked executable — download for your platform, point it at a Chromium-based browser, and off you go!

Bombadil is early but usable. Check it out on GitHub, try it on your projects, and let us know what breaks. Also let us know what it breaks in your systems! Join us on Discord for help and discussion, or follow development on Twitter/X. This is just the beginning — we’re actively seeking feedback and early adopters to help shape where this goes.