nezhar

Hi, creator of VibePod here (https://github.com/VibePod/vibepod-cli)

I spent lots of time on this topic, and had a similar path. Meanwhile the tool also supports a quick way to add custom or local providers to agents: https://vibepod.dev/news/vibepod-cli-0-24/

There is a Task mode availalbe: https://vibepod.dev/docs/agents/#managing-tasks

Workers and Boards are in the making

Nice, would https://vibepod.dev/ also be a fit for the list?

Thanks, was wondering why this is in the feed.

A good time for reading

OpenAI, OpenAPI and now OpenUI. Explain this to a non tech person...

Reminds me of https://news.ycombinator.com/item?id=42682671

Sonnet is the new haiku

This time it came with a usage reset

Yeah, same with the latest releases of meta and NVIDIA. It's like everybody is expected to have 32 GB of RAM

It feels so strange that we now have countdowns for model releases

I tried to run it with lemonade by installing it via hf but did not succeed, it gets some weird 500 errors. I also see that ollama has currently only an mlx version available.

Anybody here succeed to run this on AMD?

Sure, I added an issue for this, so it will follow in one of the next releases

You design the sandbox so the agent starts in that layer. The next thing you can do is to limit the network access, this is what I'm working on right now.

Or do you mean something else?

Open source alternative with podman support and local telemetry collection https://github.com/VibePod/vibepod-cli

Model: https://huggingface.co/mistralai/Shieldstral-1.0-3B

It's always like this, that's probably why they are pushing new models every few weeks.

Using `/model claude-opus-5[1m]` you can also use the model with older versions of Claude Code

Announcement https://hy.tencent.com/research/hy3

Free for two weeks on OpenRouter https://openrouter.ai/tencent/hy3:free

Still looking for this, the OpenKB project look promising.

I agree with the vendor lock-in aspect. My strategy was to utilize multiple agents with different APIs.

you've reached your plan's message limit

For me, building with open weights models sounds like the right approach — you are able to switch providers, and you can control where the server is running.

You don't have any guarantees in terms of data, that's true, you rely on the provider. But this is similar to a database or other services where you don't have the knowledge or resources to run them yourself. Hardware cost is an additional factor here.

If on the other hand your idea works out and the model fits the use case, you can always decide to move to a dedicated infrastructure later.

The same thing can happen in development. Data exfiltration or local file removals are often downplayed; I wonder why nobody talks about the lethal trifecta anymore.

I was just wondering if this might be the first incident. Are there any other public stories available?

It is very scary to consider the consequences that such a transfer can have.

This is, honestly, very sad.

I'm also curious about this one, also I want to compare this to vLLM.

I created VibePod, which allows you to sandbox the agents in containers and monitor their activities. It also supports OpenCode.

https://github.com/VibePod/vibepod-cli