pranshuchittora

Working on https://vostride.com/

Completely unknown, might be a very very silly leak

We believe that in future agents will collaborate with each other. And other agents would spawn a QA agent to test a hypothesis.

Yes the current agent-qa is agentic testing not a bug finder, because bug finder is mostly a token guzzler spends millions of them and it finds nothing. So we aim to solve it with the first principles, will release soon.

Good to see Momentic's demo. We at Vostride are building something similar, but we feel the verification layer should be open for verification, too.

We built agent-qa (https://github.com/vostride/agent-qa), an open-source self-improving QA agent. The harness becomes outdated every 3 to 6 months. Unfortunately, the big players are unable to maneuver their ships and are already outdated—even Momentic.

It would be great to see Momentic's benchmark on neutral benches, not the ones created by themselves.

We at Vostride have benchmarked agent-qa and are already leading on many:

- #1 on AndroidWorld benchmark (https://vostride.com/benchmark/android-world)

- AppControlBench - https://vostride.com/benchmark/app-control-bench

Checkout agent-qa at https://vostride.com/

GitHub - https://github.com/vostride/agent-qa

Many folks on x are reporting account draining

Nice coverup for the Q3 results. Wall St. don't give a F*

Thanks, looks interesting

Agree, they won't refund. But they might if the total damage is < $100K

I guess Anthropic and OpenAI might refund based on good will.

NNN... trust is already broken

Thanks for the detailed metrics. This proves my point...

Yes flagging is super downvote

Awaiting for detailed breakdown of the incident. I am more interested in knowing , how can a distillation attack leak / expose credentials.

Now it is

Yes, I am interested in how a distillation attack can leak credentials...

The tokens of frontier are valuable...

Yes

Yes, they think we are foooool

They need to explain this, by releasing a detailed RCA.

The post was visible when I am logged in. But when I opened the /newest page on incognito the post wasn't there.

It triggers me when these companies think that users are fool. I would have appreciated them, if they have disclosed things more transparently.

I am the author of this post. Looks like OpenCode fans / team is not happy with this post coming on the first page of HN. It has already been flagged by many. I never ask for upvotes, but would be great if you can upvote it. Thanks!

From OpenCode Team

---

Hi, We identified unauthorized access to a legacy system that we were in the process of decommissioning. Third party provider keys (eg OpenAI, Anthropic) saved in your workspace may have been exposed. Please revoke and replace those keys through the relevant provider’s dashboard, update any applications using them, and review recent usage for unexpected activity.

We’re sorry this happened and will follow up with any updates.

I suspect it is Vercel ;)

Yes, my Pages build is stuck in a queue for almost 2hrs

5 active incidents as of UTC 16:30 UTC

Just use EC2 t4g.nano

Andddd it broke all my EVALs... ;)

Yes, so basically all Android devices comes with Play Services that's why we need firebase. But in regions like China they have their own Play Services alternative and inorder to send notifications to those devices OneSignal or others provides a unified wrapper.

I can understand the pain as an indie dev ;(