Peter Thomas

Co-founder (YC W22) & creator of @getkarate - the only open-source tool that unifies API testing, mocks, performance and UI automation

[Peter Thomas @ptrthomas]

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

[Peter Thomas @ptrthomas]

True story. I once pitched a VC in Bangalore and he gave me Covid

[Karate @getkarate]

Karate 2.0.10 has been released. If you have been holding off on upgrading to v2, now is the time!

Check out the re-designed HTML report! Dark-mode, tag filtering and JSON syntax coloring!

Thanks to all the users already on v2 and who filed issues, we closed 25 of them in this release.

[Peter Thomas @ptrthomas]

maybe this will make your day - if you also come from the 80's and are a fan of the anime this references

https://t.co/3thrsVDTWM

[Karate @getkarate]

Code coverage tells you which lines ran.

It can't tell you which of your APIs are actually tested.

Introducing API Coverage from Karate Labs - measure coverage against the contract itself, across REST, gRPC, GraphQL & Kafka.

[Karate @getkarate]

Your API keys belong on your laptop.

Karate Xplorer: free desktop API client.

Runs your Postman collections as-is: pm.test, environments, scripts, all natively.

LLM agent built in. Bring your own model.

[Karate @getkarate]

Karate Xplorer - a free desktop API client with an LLM agent built in.

→ Zero cloud. Your keys stay local
→ Zero paywall on core features
→ Zero per-seat fees
→ Zero migration: drop in your Postman collections.

[Karate @getkarate]

A 1% error shouldn't cost you $10 Million. 📉

In the world of enterprise insurance, a single miscalculated rating factor isn’t just a "glitch"—it’s a massive drain on profitability. For a $1B premium book, a tiny 1% error translates to a $10M loss.

#InsuranceTech #InsurTech #Underwriting #RiskManagement #ActuarialScience #VeriQuant

[Karate @getkarate]

The Future of API Quality is Here.

We are proud to be featured by @Gartner_inc in the newly defined API and MCP Testing Tools category on Peer Insights.

As organizations move toward AI-native architectures, testing standard APIs is no longer enough. The need is for robust testing of the interfaces that power AI agents.

At Karate Labs, we are dedicated to providing the tools software engineering leaders need to:
🔹 Validate behavior across diverse protocols (REST, SOAP, gRPC).
🔹 Support the emerging standards of AI agent interaction.
🔹 Boost developer productivity without sacrificing quality.

Check out what the peer community is saying about us on @Gartner_Peer Insights
#TestAutomation #EngineeringExcellence #Gartner #AIArchitecture #APIManagement

[Karate @getkarate]

Announcing Karate Agent: The 100% self-hosted, air-gap ready, AI-native verification platform.

Designed specifically for highly regulated enterprise environments, Karate Agent ensures your session transcripts, screenshots, and video recordings stay securely on your own infrastructure.

🔒 Self-Hosted & Air-Gapped: Zero cloud dependency.
🧠 Bring Your Own LLM: Run with Claude, GPT, or entirely local models via Ollama. No vendor lock-in.
📹 Built-in Audit Trails: Opt-in H.264 session recording for compliance and rapid QA review.

Your proprietary application data is exactly that - yours.

Stop choosing between AI-driven efficiency and enterprise security. Reach out to our team today to schedule a demo and trial for your organization.

#EnterpriseSecurity #DataPrivacy #CyberSecurity #HealthTech #FinTech #QA #TestAutomation #KarateAgent #browseruse #rpa

[Karate @getkarate]

Karate 2.0.1 has been released with some important fixes.

https://t.co/Zfj8YcpMDq

[Karate @getkarate]

Karate 2.0 is here.

✅ One-click setup - Install the VS Code extension, hit "Karate: Setup" — done. No Java install, no Maven, no Gradle. You're running tests in under a minute.

✅ Debug JavaScript in your IDE - For the first time, you can set breakpoints and step through embedded JS in `.feature` files. Works in both VS Code and IntelliJ.

✅ Soft assertions for JSON - Match failures no longer stop your test at the first error. Every validation runs, every failure is reported. You see the full picture in one pass.

Other highlights: Native implementation of CDP and W3C WebDriver for cross-browser testing, virtual threads for parallel execution, a custom JS engine with zero dependencies, and a ground-up rewrite built on Java 21.

https://t.co/RZ0ScR5VaK

#apitesting #testautomation #cicd #performancetesting #uiautomation

[Peter Thomas @ptrthomas]

after doomscrolling I am a middle east expert. AMA

[Peter Thomas @ptrthomas]

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: https://t.co/nZ86fqISVT

[Peter Thomas @ptrthomas]

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

[Peter Thomas @ptrthomas]

@ContextQa @playwrightweb @Cypress_io yes as an enterprise add-on

https://t.co/kxIZXPDtGe

[Karate @getkarate]

We’re making a bet: The winners of this era won’t be the ones writing code the fastest, but the ones who can trust code the fastest.

Launching Karate Agent: Autonomous QA infrastructure for the Agent Era.

https://t.co/ZMy1Q9zW9l

[Peter Thomas @ptrthomas]

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."

https://t.co/RU9O61HySQ

[akira @realmcore_]

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 <Insert_Person> 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.

[Peter Thomas @ptrthomas]

In CLAUDE We Trust

[underwood @underwoodxie96]

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

[Darth Powell @VladTheInflator]

When you accidentally log onto a corporate meeting 5 minutes early