szgupta

Ah that makes sense. We could certainly investigate solutions to this. As a stopgap, if you see the log-line for the error in your output, you can copy just that log-line and ask Warp AI about that line specifically.

Hey! Currently, when you share a block (grouped input and output) with Warp AI, we let the AI know the command that was ran and also the tail of the output. (and to streamline this even more, sharing a block with Warp AI is as simple as one-click!)

Often times, when these longer scripts / build jobs fail, they'll emit an error at the _end_, so sharing the tail should work in this case. But, you can always tweak (add / subtract) from what you want to share with the AI to make the query even more focused.

Does that address your use-case?

Hey, thanks for the suggestion! We have plans to further integrate the shell-context so that the AI can curate its answers even better. For example, letting the AI know that you're working in bash vs. fish so the commands it suggests respect syntax, etc.

Currently, you can share a Warp block with the AI, which will allow the AI to see both your command and the tail of your output. You can also right-click on any text selection and ask Warp AI -- this would also be a good place for us to attach the top-level command you're running (e.g. psql, gdb, vim) to the context. I'll note that down. Would that address your use-case?

Hi HN, I’m Suraj, an engineer at Warp (www.warp.dev) where we’re building a fast, Rust-based terminal for the 21st century. Today I am excited to share Warp AI with the community.

With this launch we have added AI to the core interface of the terminal. This integration allows you to ask Warp AI to explain command line errors and output, execute suggested fixes, walk you through complex CLI workflows, or write entire scripts for you.

For example, you can ask Warp AI to help you tail logs from a gcloud instance, generate new ssh certs, or run a suggested fix, e.g. ps aux | grep 'python' | awk '{print $2}' | xargs kill -9.

Warp AI also includes AI Command Search, which lets you generate any command using natural language — especially helpful for commands you know but can’t remember at the moment, saving you a trip to google or stack overflow.

All of this is built into the terminal interface itself which has benefits. There’s no need to run a separate CLI to access the AI interface. It lives within the terminal. You can run suggested commands directly from the AI response with the click of a button.

Here’s our official blog post announcing the feature: www.warp.dev/blog/introducing-warp-ai

Here’s a video of the feature in action: https://www.youtube.com/watch?v=1YZLPJ5X18k

A few notes on privacy and security: Telemetry in Warp — and Warp AI — is completely opt-in. The data we send is transparently documented [1] for reference. You can inspect all network requests made by Warp via our network log feature [2].

Also, when using Warp AI, no data from inputs is stored on Warp servers. The integration is built on OpenAI. Warp does not store any data, but rather proxies the requests.

It’s also important to call out: A login is required to use both Warp and Warp AI, similar to other cloud-based services.

Warp AI is in a free preview for developers on Mac, up to 100 requests per day.

Please try it out and let us know what you think!

[1] https://docs.warp.dev/getting-started/privacy#exhaustive-tel...

[2] https://docs.warp.dev/features/network-log

Hi HN, I’m Suraj, an engineer at Warp (www.warp.dev) where we’re building a fast, Rust-based terminal for the 21st century. Today I am excited to share Warp AI with the community.

With this launch we have added AI to the core interface of the terminal. This integration allows you to ask Warp AI to explain command line errors and output, execute suggested fixes, walk you through complex CLI workflows, or write entire scripts for you.

For example, you can ask Warp AI to help you tail logs from a gcloud instance, generate new ssh certs, or run a suggested fix, e.g. ps aux | grep 'python' | awk '{print $2}' | xargs kill -9.

Warp AI also includes AI Command Search, which lets you generate any command using natural language — especially helpful for commands you know but can’t remember at the moment, saving you a trip to google or stack overflow.

All of this is built into the terminal interface itself which has benefits. There’s no need to run a separate CLI to access the AI interface. It lives within the terminal. You can run suggested commands directly from the AI response with the click of a button.

Here’s our official blog post announcing the feature: www.warp.dev/blog/introducing-warp-ai

Here’s a video of the feature in action: https://www.youtube.com/watch?v=1YZLPJ5X18k

A few notes on privacy and security: Telemetry in Warp — and Warp AI — is completely opt-in. The data we send is transparently documented [1] for reference. You can inspect all network requests made by Warp via our network log feature [2].

Also, when using Warp AI, no data from inputs is stored on Warp servers. The integration is built on OpenAI. Warp does not store any data, but rather proxies the requests.

It’s also important to call out: A login is required to use both Warp and Warp AI, similar to other cloud-based services.

Warp AI is in a free preview for developers on Mac, up to 100 requests per day.

Please try it out and let us know what you think!

[1] https://docs.warp.dev/getting-started/privacy#exhaustive-tel...

[2] https://docs.warp.dev/features/network-log

I love this sentiment. Your questions would make for a very interesting post.

While we haven't rebuilt from absolute ground zero, Warp is definitely trying to extend the capabilities of a terminal (emulator) from what's historically been possible. For example, we introduced a dedicated input editor so you can have an IDE-like experience in the terminal. It's fundamentally different from how input is entered in a traditional terminal. But with this innovation, we've had to be careful to ensure that all the input features you expect in a normal terminal (even obscure ones like `alt-.`) work how you'd expect, _and then some_.

Overall though, starting from scratch is hard because we need to stay backwards-compatible with all the CLIs we use everyday.

Thanks for the comment. I definitely feel your concerns.

To clarify, all cloud-oriented features are fully opt-in. For example, you have to explicitly share a block for us to store it in the cloud.

You can also opt-out of telemetry (which we use to determine feature usage and plan our roadmap FWIW).

We even have a network log so you can see every network call we make.

More details here: https://www.warp.dev/blog/telemetry-now-optional-in-warp

Yeah I like to think of it as a capability list as well. There was a brief footnote [12] about $TERM in the post, but terminfo could certainly be dissected further.

[12](https://www.warp.dev/blog/what-happens-when-you-open-a-termi...)

Hey, sorry about that! We actually have a WIP fix for this exact issue (the obtrusive overlay that you can't dismiss). Should be out in the next week or two.

Haha, yeah we probably should have added a caveat that this is only applicable to Unix

Haha! The post was actually motivated by a common (interview) question: https://github.com/alex/what-happens-when. We thought it would make for a fun adaptation for the terminal world

Thanks! It was fun to write this up. Glad you enjoyed :)

Thanks! Were there any topics in the blog that you think would benefit from more detail in a follow-up post?

Hey HN! I'm Suraj, one of the authors of this blog post. Andy (@acarl005) and I came up with the idea to write this post when we realized there weren't many existing resources on how a terminal works under the hood, end-to-end. If folks find this useful, we'd be happy to turn this into a series and dive deeper into sub-topics. Let us know what you think :)