Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Cursor CLI

cursor.com

Post reply on HN
Print view View original

81–90 of 275 posts

Re: Cursor CLI

#81

Post by ribeyes »

i'm betting on cursor being the long-term best toolset.

1. with tight integration between cli, background agent, ide, github apps (e.g. bugbot), cursor will accommodate the end-to-end developer experience.

2. as frontier models internalize task routing, there won't be much that feels special about claude code anymore.

3. we should always promote low switching costs between model providers (by supporting independent companies), keeping incentives toward improving the models not ui/data/network lock-in.

Re: Cursor CLI

#82

Post by ankit219 »

I think CLI is a good idea for now. Next abstraction seems to be Github PRs where someone (likely me) files an issue/feature, then I click a button, and the agent fixes the issue/feature. Github has talked about something similar, but surely it were a pain to figure out if it was GA and I had access to it given so many different variations they have called gh copilot. (PS: it exists, but not as smooth as I described: https://docs.github.com/en/copilot/how-tos/use-copilot-agent... )

Re: Cursor CLI

#84

Post by ramoz »

ribeyes wrote: post #81

i'm betting on cursor being the long-term best toolset. 1. with tight integration between cli, background agent, ide, github apps (e.g. bugbot), cursor will accommodate the end-to-end developer experience. 2. as frontier models internalize task routing, there won't be much that feels special about claude code anymore. 3. we should always promote low switching costs between model providers (by supporting independent c…

Happy to short that bet as I think agentic harnesses will be molded along the RL training of the actual model. Tony + the suit created together. Why Claude in Claude Code became existential for Cursor, why cursor moved quick to go agentic and build up with OpenAI in big header line way here.

Unless they pair up with OpenAI or Meta.

Re: Cursor CLI

#85

Post by byronic »

I'm having trouble finding a use for this outside of virtualized unused environments. Why not instead give me a virtual machine that runs this in a confined storage space?

I would _never_ give an LLM access to any disk I own or control if it had anything more than read permissions

Re: Cursor CLI

#88

Post by postalcoder »

ribeyes wrote: post #81

i'm betting on cursor being the long-term best toolset. 1. with tight integration between cli, background agent, ide, github apps (e.g. bugbot), cursor will accommodate the end-to-end developer experience. 2. as frontier models internalize task routing, there won't be much that feels special about claude code anymore. 3. we should always promote low switching costs between model providers (by supporting independent c…

i’d respectfully bet against this.

cursor and 3rd party tools will, unless they make their own superior foundation model, will always have to fight the higher marginal cost battle. This is particularly bad insofar that they offer fixed pricing subscriptions. That means they’re going to have to employ more context saving tricks which are at odds with better performance.

If the cost economics result in Cursor holding, say, 20% fewer tokens in context versus model-provider coding agents, they will necessarily get worse performance, all things equal.

Unless Cursor offers something dramatically different outside of the basic agentic coding stack it’s hard to see why the market will converge to cursor.

Re: Cursor CLI

#89

Post by ximeng »

thehamkercat wrote: post #62
prmph wrote: post #55

Earlier quoted context omitted.

Yep, that's a peeve of mine. I've resorted to using AGENT.md, and aliasing Claude, Gemini, etc to a command that calls them with an initial instruction to read that file. But of course they will forget after some time. The whole agentic coding via CLI experience could be much improved by: - Making it easy to see what command I last issued, without having to scroll up through reams of output hunting for context - Maki…

I think most of them provide an option to change the default file, but it'll be really good if they all can switch to AGENT.md by default Till then you can also use symlinks there are issues opened in some repos for this - Support "AGENT.md" spec + filename · Issue #4970 · google-gemini/gemini-cli https://github.com/google-gemini/gemini-cli/issues/4970#issu...

https://github.com/anthropics/claude-code/issues/1091

Here for Claude

Re: Cursor CLI

#90

Post by dgunay »

thehamkercat wrote: post #21

I wonder when all of them will adopt AGENT.md and stop using gemini.md/claude.md/crush.md/summary.md/qwen.md https://agent.md [redirect -> https://ampcode.com/AGENT.md ] https://agent-rules.org

I just wish the AGENTS.md standard wasn't a single file. I have a lot of smaller context documents that aren't applicable to every task, so I like to throw them into a folder (.ai/ or .agents/) and then selectively cat them together or tell the agent to read them.
Post reply on HN