Peter Thomas

In case you missed the Karate v2 announcement, link is in the comments.

after doomscrolling I am a middle east expert. AMA

instead of full blown MCP + json-schema approach, I suggest claude code (or eq) just writes pure python over HTTP, much faster, full control. sounds like you may have figured this out already, and your MCP tool is just an envelope over the blender python API. here's my initial version, already able to do a lot, and quite fast ! planning to automate the video sequence editor because I need it: github.com/ptrthomas/blen…

@em0tionull It built its own blender mcp and runs in an agentic loop, improving its techniques and tools autonomously

I built an agentic system that taught itself the Blender donut tutorial by watching it on YouTube. It watched the tutorials, extracted the steps, filled in the gaps in own tooling and completed the entire thing autonomously.

ah so that's why there weren't any fireworks last night at unearthly hours. serves you right

@ContextQa @playwrightweb @Cypress_io yes as an enterprise add-on x.com/getkarate/stat…

@ptrthomas @playwrightweb @Cypress_io Karate is an impressive framework for API testing. The debate about tool choice often overshadows the maintenance burden. Self-healing tests could complement Karate's capabilities. Are you exploring auto-maintenance features?

Five reasons why @playwrightweb is better than @Cypress_io "I just think it’s incorrectly marketed as an end-to-end testing tool when it’s really only good for component testing." alisterbscott.com/2021/10/27/fiv…

Feels like everyone making their own agent stumbles across the same primitives and thinks they solved something Let me save you some time (read this, it's funny and useful): - You're going to make an agent - You're going to run it on benchmarks > It's going to suck - You're going to make a tool to analyze traces - You're going to say this helped you > It wont work - You're going to think about role based agents for solving a single task - You're going to make a workflow for solving a benchmark > Both will work. Neither are generalized - You'll think you made it > It will be nearly unusable by an end user > Back to square one - You're going to realize you're stuck with a for loop - You're going to think about swarms > In swarms single agent usability doesn't matter - But wait you need a task manager - But wait you need a merge queue - But wait you need compression for long jobs > Compression is a foot gun - But wait now you need an agent to manage it all - But wait now you need something that checks to make sure the manager is managing - You're going to go back to single loop agents - Well, subagents seem like the way to do all of this > Bam! Plot twist: subagents are hard to do well - You're going to think "Hmm well subagents isolate context" because said so - You're going to start to look at other agent implementations > How have they all solved compaction, multi-agent, task management, memory etc.? - You're going to realize it's all just tradeoffs, but most of them have only one side people care about - "Oh it's all just context engineering" > Yep. But it has to be good and it has to be general. > Back to the starting loop. Rinse and repeat. Congrats. Keep it simple. Keep it general.

In CLAUDE We Trust

Thanks to seedance 2.0, They were relieved of their suffering.

When you accidentally log onto a corporate meeting 5 minutes early

Test Automation, Reimagined - Karate v2 is here! ==================================== Following-up on our original announcement, we're excited to announce that Karate v2 is now available on Maven Central for you to try! Read on to see what's new...

impressed at how this is articulated. the best engineers can write and explain things.

@maxandersen ok lemme guess you are going to contribute to one or write your own ;)

@ptrthomas I don't have a strong recommendation beyond what I linked to in the blog - they aren't "great" - yet. Ill share more in early 2026 :)

Welcoming Mamie F Jones to Karate Labs Advisory Board. Karate is used by 650+ companies, including 76 of the F500. With marquee brand names as customers, we aspire to be the top solution for API testing across Banking, Insurance, Financial Services, Retail and Software. In May 2024, we welcomed Zachary Griesbach from Guidewire Software to our Advisory Board to provide industry-specific guidance for the Insurance vertical. Today, we're thrilled to add expertise in organizational transformation and leadership excellence as we continue to scale Karate Labs to a category-defining enterprise solution. It is our pleasure to have Mamie F. Jones join Karate Labs Advisory Board. Mamie brings 35+ years of transformation leadership across technology giants including Intuit, Sabre, Travelocity, Dun & Bradstreet, and Dell. As former SVP of Product Development at Intuit, she led the company's cloud and platform transformation, reducing operational costs by 25% while accelerating cloud migration in under two years. She spearheaded innovative ML initiatives that revolutionized the accounting and tax industry. Mamie is the author of "Lead Extraordinary Change: The Proven Playbook for Driving Successful Organizational Transformation" and has received numerous honors including the CEO Leadership Award at Intuit (2016) and the Tech Titan CTO Award for North Texas (2015). She currently serves as an Independent Board Director at Mastek Inc. While Zachary guides us on Insurance industry requirements and go-to-market strategy, Mamie will focus on organizational transformation, building high-performing engineering teams, fostering innovation culture, operational excellence, and strategic growth as we scale across all our target verticals. Mamie, welcome to the Karate Labs Advisory Board. We look forward to your strategic guidance, mentoring and having you part of our journey.

@BideshiMod thanks, you are right

@ptrthomas if i saw it correctly, she was a developer advocate and not the product owner? nevertheless her content was so good and helpful.

just saw Microsoft laid off the product owner for Playwright ! honestly I think that's a shame, she was doing a kick ass job quoting a tweet from the past as a grim reminder. test automation tools are not core to companies like MS

@simonw because it is Textual does it feel snappier than Claude Code ?

Here are pelicans for Mistral's two new models, Devstral 2 and Devstral Small 2 - I'm quite impressed given the smaller parameter size of these models (123B and 24B) simonwillison.net/2025/Dec/9/dev…

Put together some notes on Mistral Vibe, Mistral's new CLI coding agent (think Claude Code / OpenAI Codex) which is Apache 2 licensed, written in Python+Pydantic+Textual and has a neat set of system and tool definition prompts that are fun to read simonwillison.net/2025/Dec/9/mis…

well that is interesting. I was really hoping for an alternative to VS Code to succeed tempted to start an OSS project for a Java based IDE platform ...

@simonw may be off topic, but sharing in case you have not seen this resource: github.com/OWASP/www-proj…

Something I really want to exist is a comprehensive guide to the sandboxing feature of all of the popular coding agents: what they do, how they work, how reliable they are (I have a nasty feeling I'm destined to pull this together myself and I really don't want to do the work)

Karate brings API test automation, performance checks, and UI testing into one powerful framework, helping teams deliver faster with cleaner, readable scenarios. Learn why it stands out and how you can get started with building your first tests in Ahmet Geçer's blog from our QA team. 🔗 commencis.com/thoughts/karat… #APITesting #Automation #KarateFramework #SoftwareTesting #QATeam

In case you missed the Karate v2 announcement, link is in the comments.