[HN Gopher] Show HN: Aide, an open-source AI native IDE
___________________________________________________________________
Show HN: Aide, an open-source AI native IDE
Hey HN, We are Sandeep and Naresh, the creators of Aide. We are
happy to open source and invite the community to try out Aide which
is a VSCode fork built with LLMs integrated. To talk through the
features, we engineered the following: - A proactive agent Agent
which iterates on the linter errors (powered by the Language
Server) and pulls in relevant context by doing go-to-definitions,
go-to-references etc and propose fixes or ask for more files which
might be missing in the context. - Developer control We encourage
you to do edits on top of your coding sessions. To enable this, we
built a VSCode native rollback feature which gets rid of all the
edits made by the agent in a single click if there were mistakes,
without messing up your changes from before. - A combined
chat+edit flow which you can use to brainstorm and edit You can
brainstorm a problem in chat by @'ting the files and then jump into
edits (which can happen across multiple files) or go from a smaller
set of edits and discuss the side-effects of it - Inline editing
widget We took inspiration from the macos spotlight widget and
created a similar one inside the editor, you can highlight part of
the code, do Cmd+K and just give your instructions freely - Local
running AI brain We ship a binary called sidecar which takes care
of talking to the LLM providers, preparing the prompts and using
the editor for the LLM. All of this is local first and you get full
control over the prompts/responses without anything leaking to our
end (unless you choose to use your subscription and share the data
with us) We spent the last 15 months learning about the internals
of VSCode (its a non-trivial codebase) and also powering up our AI
game, the framework is also at the top of swebench-lite with 43%
score. On top of this, since the whole AI side of the logic runs
locally on your machine you have complete control over the data,
from the prompt to the responses and you can use your own API Keys
as well (can be any LLM provider) and talk to them directly.
There's still a whole lot to build and we are at 1% of the journey.
Right now the editor feels robust and does not break on any of the
flows which we aimed to solve for. Let us know if there's anything
else you would like to see us build. We also want to empower
extensibility and work together with the community to build the
next set of features and set a new milestone of AI native editors.
Author : skp1995
Score : 236 points
Date : 2024-11-06 15:01 UTC (1 days ago)
(HTM) web link (aide.dev)
(TXT) w3m dump (aide.dev)
| morgante wrote:
| Is the "sidecar" open source too?
| ghostwriternr wrote:
| Yes, it is! https://github.com/codestoryai/sidecar
| WillAdams wrote:
| Confusingly, "Sidecar" is the name Apple uses for their
| feature of having an iPad serve as a second screen/touch
| interface for a Mac:
|
| https://support.apple.com/en-us/102597
| Validark wrote:
| I believe I get the metaphor. Why is it confusing?
| WillAdams wrote:
| Overloading the term with a second technological meaning.
| scubbo wrote:
| Unfortunately, there are a distinctly limited number of
| words that can communicate a particular concept - the
| pigeon-hole principle suggests duplication will be
| inevitable.
|
| Ambiguity's only really a problem when the same term is
| used multiple ways in similar contexts. I think it's very
| unlikely that anyone will get confused between these two
| usages.
| yjftsjthsd-h wrote:
| It's also the name of a k8s thing: https://kubernetes.io/do
| cs/concepts/workloads/pods/sidecar-c...
| skp1995 wrote:
| In my previous life at Facebook, I worked on the infra
| team and worked on a cluster manager similar to
| kubernetes, thats where I first heard the term sidecar.
| Something about the concept of a binary running alongside
| the pod powering other related things felt strong. In
| most parts this is the inspiration for naming the AI
| brain: sidecar
| rompledorph wrote:
| And AIDE is also the name for an Android IDE or an
| Advanced Intrusion Detection Environment
| ilrwbwrkhv wrote:
| So what? Just because Apple calls something retina
| "display" means that others cannot call stuff displays.
| homarp wrote:
| read it as sAIDEcar
| pseudo_rand000 wrote:
| What differentiates Aide from all the existing tools in this
| space like Cursor?
| skp1995 wrote:
| VSCode forks are not new, there are many companies out there
| building towards this vision. What sets us apart is partly our
| philosophy (deeply integrating into the editor) and also the
| tech stack (running everything to the dot locally) and giving
| developers control over the LLM usage and also other niceness
| (like rollbacks which I think are paramount and important)
| swyx wrote:
| > What sets us apart is partly our philosophy (deeply
| integrating into the editor)
|
| i'm so sorry but what do you think cursor's philosophy is
|
| > also other niceness (like rollbacks
|
| yep in cursor too
|
| i know youre new so just being gentle but try to focus on
| some kind of killer feature (which i guess is sidecar?)
|
| also https://x.com/codestoryai seems suspended fyi
| Alifatisk wrote:
| > I'm so sorry but what do you think cursor's philosophy is
|
| I've never understood why people say sorry for cases like
| these?
| aniviacat wrote:
| Because without the sorry the comment may be interpreted
| in an aggressive/dismissive tone. For this issue, adding
| a sorry is an easy fix.
| scubbo wrote:
| Can be a cultural idiomatic thing. For me (BrEng) "I'm
| sorry but..." doesn't explicitly mean "I regret/apologise
| for what I'm about to say" - in fact it's an intensifier.
| skp1995 wrote:
| Fair point! We are not taking a stab at cursor in any way
| (its a great product)
|
| In terms of features I do believe we are differentiated
| enough, the workflows we came up with are different and we
| are all about giving back the data (prompt+responses) back
| to the user.
|
| The sidecar is not the killer feature, its one of the many
| thing which ticks the whole experience together.
|
| Good callout on the codestoryai account being suspended, we
| are at @aide_dev
| swyx wrote:
| you link to it still on your home page
| skp1995 wrote:
| great catch! Thank you for pointing this out
| joenot443 wrote:
| Feature wise, what are you offering I wouldn't get in Cursor
| today?
| OsrsNeedsf2P wrote:
| Aide seems to have a good open source license (Cursor is
| proprietary)
| skp1995 wrote:
| Open Source; giving full ownership of the data to the users;
| running completely locally; we want to make sure you can use
| Aide no matter the environment you are in.
| xpasky wrote:
| Any short-term plans for Claude via AWS Bedrock? (That's for me
| personally a blocker for trying it on our main codebase.)
| skp1995 wrote:
| Thanks for your interest in Aide!
|
| If I understood that correctly, it would mean supporting Claude
| via the AWS Bedrock endpoint, we will make that happen.
|
| If the underlying LLM does not change then adding more
| connectors is pretty easy, I will ping the thread with updates
| on this.
| xpasky wrote:
| Yep! And AWS Bedrock gives you also plenty of other models on
| the back end, plus better control over rate limits. (But for
| us the important thing is data residency, the code isn't
| uploaded anywhere.)
|
| Is it ~just about adding another file to https://github.com/c
| odestoryai/sidecar/blob/main/llm_client/... ?
|
| I could take a look too - another way for me to test Aide by
| working with it to implement this. :-)
|
| (https://github.com/pasky/claude.vim/blob/main/plugin/claude_
| ... is sample code with basic wrapper emulating Claude
| streaming API with AWS Bedrock backend.)
| skp1995 wrote:
| yup! feel free to add the client support, you are on the
| right track with the changes.
|
| To test the whole flow out here are a few things you will
| want to do: - https://github.com/codestoryai/sidecar/blob/b
| a20fb3596c71186... (you need to create the LLMProperties
| object over here) - add support for it in the broker over
| here: https://github.com/codestoryai/sidecar/blob/ba20fb359
| 6c71186... - after this you should be at the very least
| able to test out Cmd+K (highlight and ask it to edit a
| section) - In Aide, if you go to User Settings: "aide self
| run" you can tick this and then run your local sidecar so
| you are hitting the right binary (kill the binary running
| on 42424 port, thats the webserver binary that ships along
| with the editor)
|
| If all of this sounds like a lot, you can just add the
| client and I can also take care of the plumbing!
| xpasky wrote:
| Hmm looks like this is still pretty early project for me.
| :)
|
| My experience: 1. I didn't have a working installation
| window after opening it for the first time. Maybe what
| fixed it was downloading and opening some random
| javascript repo, but maybe it was rather switching to
| "Trusted mode" (which makes me a bit nervous but ok).
|
| 2. Once the assistant window input became active, I wrote
| something short like "hi", but nothing happenned after
| pressing ctrl-Enter. I rageclicked around a bit, it's
| possible I have queued multiple requests. About 30
| seconds later, suddenly I got a reply (something like "hi
| what do you want me to do"). That's .. not great latency.
| :)
|
| 3. Since I got it working, I opened the sidecar project
| and sent my _second_ assistant prompt. I got back this
| response after few tens of seconds: "You have used up
| your 5 free requests. Please log in for unlimited
| requests." (Idk what these 5 requests were...)
|
| I gave it one more go by creating an account. However
| after logging in through the browser popup, "Signing in
| to CodeStory..." spins for a long time, then disappears
| but AIDE still isn't logged in. (Even after trying again
| after a restart.)
|
| One more thought is maybe you got DDos'd by HN?
| skp1995 wrote:
| > 2. Once the assistant window input became active, I
| wrote something short like "hi", but nothing happenned
| after pressing ctrl-Enter. I rageclicked around a bit,
| it's possible I have queued multiple requests. About 30
| seconds later, suddenly I got a reply (something like "hi
| what do you want me to do"). That's .. not great latency.
| :)
|
| Yup thats cause of the traffic and the LLM rate limits :(
| we are getting more TPM right now so the latency spikes
| should go away, I had half a mind to spin up multiple
| accounts to get higher TPM but oh well.... if you do end
| up using your own API Key, then there is no latency at
| all, right now the requests get pulled in a global queue
| so thats probably whats happening.
|
| > 3. Since I got it working, I opened the sidecar project
| and sent my second assistant prompt. I got back this
| response after few tens of seconds: "You have used up
| your 5 free requests. Please log in for unlimited
| requests." (Idk what these 5 requests were...)
|
| The auth flow being wonky is on us, we did fuzzy test it
| a bit but as with any software it slipped from the
| cracks. We were even wondering to skip the auth
| completely if you are using your own API Keys, that way
| there is 0 touch interaction with our llm proxy infra.
|
| Thanks for the feedback tho, I appreciate it and we will
| do better
| solarkraft wrote:
| Not only is the name Aide already used by another project, it's
| even also an IDE.
|
| https://www.android-ide.com/
| skp1995 wrote:
| TIL, I thought we covered the ground when grepping for Aide.
| Funny that its also an editor
| memsom wrote:
| It is a pretty well established IDE. I used it back on a
| Nexus 4 when that phone was actually "recent" to give you
| some context.
| skp1995 wrote:
| I do remember the Nexus 4 (jelly bean OS). I was fascinated
| at the time that you could play games on Android and ran
| the emulator for android on my desktop at that point (I was
| young and needed the games haha)
| Mashimo wrote:
| Looks rather dead.
| fermigier wrote:
| AIDE has been around for 25 years: https://aide.github.io/
|
| IMHO the right thing would be to use another name.
| skp1995 wrote:
| I ... did not know that.
|
| We should probably pick another name then
| dsabanin wrote:
| You probably shouldn't.
| lioeters wrote:
| The name is perfect, AI + IDE = Aide. You should keep it.
| pabs3 wrote:
| Maybe add a hyphen; ai-ide
| handfuloflight wrote:
| There's also https://aider.chat/, which is... close.
| MyFedora wrote:
| This is literally a totally different piece of software with
| a completely unrelated use case. Changing the name would make
| as much sense as renaming a hammer because someone invented a
| screwdriver.
| mellosouls wrote:
| Links to the project, I'm guessing these :)
|
| https://github.com/codestoryai/aide
|
| https://aide.dev/
| skp1995 wrote:
| you missed this one https://github.com/codestoryai/sidecar
| Sidecar: The AI brains Aide:
| https://github.com/codestoryai/aide the editor
| unshavedyak wrote:
| Any tips for using Aide with another text editor? Ie i'm not
| going to work outside of my preferred text editor (Helix atm), so
| i'm curious about software which has a workflow around this.
| Rather than trying to move me to a new text editor
| skp1995 wrote:
| hmm.... I do think it can be extended to work outside of just
| the VSCode environment.
|
| If you look at the sidecar side of things:
| https://github.com/codestoryai/sidecar/blob/ba20fb3596c71186...
| these are the main APIs we use
|
| On the editor side:
| https://github.com/codestoryai/ide/blob/0eb311b7e4d7d63676ad...
|
| These are the access points we need
|
| The binary is fairly agnostic to the environment, so there is a
| possibility to make it work elsewhere. Its a bit non trivial
| but I would be happy to brainstorm and talk more about this
| swlkr wrote:
| I also use helix and i've been getting some mileage out of
| aider, the cli tool. Confusing name, as I don't believe aider
| is affiliated with aide
| skp1995 wrote:
| do you know of helix exposes the LSP APIs all the way to the
| editor .. if it does doing the integration should be trivial
| swlkr wrote:
| im fairly sure helix-gpt does this, though i haven't tried
| it
| skp1995 wrote:
| reading the code for helix-gpt over here
| https://github.com/leona/helix-
| gpt/blob/2a047347968e63ca55e2... looks like the
| architecture for the extension is based around getting
| the diagnostic event and then passing that along to the
| chat.
|
| The readme also talks about how LSP services are not
| exposed properly yet, my takeaway is that its not
| complete yet..but surely doable
| drag0s wrote:
| Looks like the download links from your landing page are broken?
| Looks like our build pipeline is broken! Click here to let
| us know?
| skp1995 wrote:
| wooops... on it (we got rate limited by Github) in the
| meanwhile
| https://github.com/codestoryai/binaries/releases/tag/1.94.2....
| check this out
| skp1995 wrote:
| its fixed!
| renewiltord wrote:
| You have just woken up from the cryosleep you entered in 2024.
| The year is 2237. GPT-64 and its predecessors have been around
| for nigh on 100 years. But there has been no civilizational
| upheaval. Your confusion is cleared when you check the inter-
| agent high-speed data bus. You expect this to be utterly
| incomprehensible, but both the human and AI data is clearly
| visible. It is a repeating pattern. The agents are mimicking
| human behavior perfectly and you can't tell which is which. All
| data transmitted has the same form: $word is
| already a name for a project. Stop copying it. Change your name.
|
| Mankind and His Machine Children have met The Great Filter.
| skp1995 wrote:
| hahaha
| __MatrixMan__ wrote:
| I don't think it's going to take us 200 years to kick the habit
| of using global namespaces for friendly names, maybe 80.
| Recognizing a name and rendering it as a disambiguation based
| on my location in the trust graph should be a feature of the
| text box, not something that I have to think about.
| DesiLurker wrote:
| sounds like the scene in movie Idiocracy where roomba is stuck
| in corner and keeps repeating 'floor is now clean'.
| Alifatisk wrote:
| Aide.dev is similar to aider.chat except Aide being an IDE while
| Aider is a CLI
| skp1995 wrote:
| AIDE == AI + IDE (that was our take on the name)
| james_marks wrote:
| Looks interesting, is there a binary for Mac OS? I'd rather not
| build from scratch just to demo.
|
| For the people comparing to Cursor on features, I suspect the
| winner is going to be hard to articulate in an A:B comparison.
|
| There's such a difference in feel that may be rooted in a
| philosophy, but boils down to how much the creator's vision
| aligns with my own.
| skp1995 wrote:
| Yes there is, we have the binary link on our website but
| putting it here:
|
| - arm64 build:
| https://github.com/codestoryai/binaries/releases/download/1....
|
| - x86 build:
| https://github.com/codestoryai/binaries/releases/download/1....
|
| > There's such a difference in feel that may be rooted in a
| philosophy, but boils down to how much the creator's vision
| aligns with my own.
|
| Hard agree! I do think AI will find its way into our
| productivity tool kit in different ways. There are still so
| many ways we can go about doing this, A:B comparison aside I do
| feel the giving people to power to mold the tool to work for
| themselves is the right way.
| h1fra wrote:
| Genuine question, with vscode going all-in in this direction,
| what's left for forks like this?
| skp1995 wrote:
| There are quite a few things! VSCode's direction (I am making
| my own assumptions from the learnings I have) - VSCode is
| working on the working set direction of making multi-file edits
| work - Their idea of bringing in other extension is via the
| provider API which only copilot has access to (so you can't use
| them if you are not a copilot subscriber)
|
| So just taking these things for face value, I think there is
| lots to innovate. No editor (bias view of mine) has really
| captured the idea of a pair programmer working alongside you.
| Even now the most beloved feature is copilot or cursor tab with
| the inline completions.
|
| So we are ways further from a saturated market or even a
| feature set level saturation. Until we get there, I do think
| forks have a way to work towards their own version of an ideal
| AI native editor, I do think the editors of the future will
| look different given the upwards trend of AI abilities.
| rubslopes wrote:
| > No editor (bias view of mine) has really captured the idea
| of a pair programmer working alongside you.
|
| I had this feeling for the first time with Cline. It adjusted
| the code, accessed the terminal, rebuilt the image, ran the
| container, saw an error, suggested new edits, ran again...
| All while only asking for a few confirmations. And it's very
| verbose: it tells you with details what it's doing every step
| of the way.
|
| But I migrated to the new Copilot a few days ago because I
| was easily spending $5 in a day.
| arjunaaqa wrote:
| Betting on Microsoft messing up on UX side, as always.
| hubraumhugo wrote:
| I'm curious - what does the AI coding setup of the HN community
| look like, and how has your experience been so far?
|
| I want to get some broader feedback before completely switching
| my workflow to Aide or Cursor.
| arjunaaqa wrote:
| Using cursor and it's been great !
|
| Founders care about development experience a lot and it shows.
|
| Yet to try others, but already satisfied so not required.
| skp1995 wrote:
| I can give my broader feedback: - Codegen tools today are still
| not great: The lack of context and not using LSP really burns
| down the quality of the generated code. - Autocomplete is great
| Autocomplete is pretty nice, IMHO it helps finish your thoughts
| and code faster, its like intellisense but better.
|
| If you are working on a greenfield project, AI codegen really
| shines today and there are many tools in the market for that.
|
| With Aide, we wanted it to work for engineers who spend >= 6
| months on the same project and there are deep dependencies
| between classes/files and the project overall.
|
| For quick answers, I have a renewed habit of going to
| o1-preview or sonnet3.5 and then fact checking that with google
| (not been to stack overflow in a long while now)
|
| Do give AI coding a chance, I think you will be excited to say
| the least for the coming future and develop habits on how to
| best use the tool.
| SparkyMcUnicorn wrote:
| > Codegen tools today are still not great: The lack of
| context and not using LSP really burns down the quality of
| the generated code
|
| Have you tried Aider?
|
| They've done some discovery on this subject, and it's
| currently using tree-sitter.
| skp1995 wrote:
| Yup, I have.
|
| We also use tree-sitter for the smartness of understanding
| symbols https://github.com/codestoryai/sidecar/blob/ba20fb3
| 596c71186... and also the editor for talking to the
| Language Server.
|
| What we found was that its not just about having access to
| these tools but to smartly perform the `go-to-definition`
| `go-to-reference` etc to grab the right context as and when
| required.
|
| Every LLM call in between slows down the response time so
| there are a fair bit of heuristics which we use today to
| sidestep that process.
| tomr75 wrote:
| cursor works well - uses RAG on your code to give context, can
| directly reference latest docs of whatever you're using
|
| not perfect but good to incrementally build things/find bugs
| nprateem wrote:
| I tried GH copilot again recently with Claude. It was complete
| shit. Dog slow and gave incomplete responses. Back to aider.
| skp1995 wrote:
| what was so bad about it? genuinely curious cause they did
| make quite a bit of noise about the integration.
| nprateem wrote:
| It kept truncating files only about 600 lines long. It also
| seems to rewrite the entire file each time instead of just
| sending diffs like aider making it super slow.
| skp1995 wrote:
| oh, I see your point now. Its weird that they are not
| doing the search and replace style editing. Altho now
| that OpenAI also has Predicted Output, I think this will
| improve and it won't make mistakes while rewriting longer
| files.
|
| The 600 line limit might be due to the output token limit
| on the LLM (not sure what they are using for the code
| rewriting)
| nprateem wrote:
| Yeah I guess it's a response limit. It makes it a deal
| breaker though.
| csomar wrote:
| Interestingly, I had this problem with Claude (their web
| chat) and not Copilot. However, there were times where it
| was unresponsive.
| HyprMusic wrote:
| It's not nearly as helpful as Claude.ai - it seems to only
| want to do the minimum required. On top of that it will
| quite regularly ignore what you've asked, give you back the
| exact code you gave it, or even generate syntactically
| invalid code.
|
| It's amazing how much difference the prompt must make
| because using it is like going back to gpt3.5 yet it's the
| same model.
| xpasky wrote:
| Besides Claude.vim for "AI pair programming"? :) (tbh it works
| well only for small things)
|
| I'm using Codeium and it's pretty decent at picking up the
| right context automatically, usually it autocompletes within
| ~100kLoC project quite flawlessly. (So far I haven't been using
| the chat much, just autocomplete.)
| skp1995 wrote:
| any reason you don't use the chat often, or maybe it's not
| your usecase?
| Ancapistani wrote:
| I'm not the parent poster, but in my case I very rarely use
| it because it's not in the Neovim UI; it opens in a
| browser.
|
| I've also had some issues where it doesn't seem to work
| reliably, but that could be related to my setup.
| skp1995 wrote:
| yeah I am learning that on neovim you can own a buffer
| region and instead use that for ai back and forth.. it's
| a very interesting space
| viraptor wrote:
| Cursor works amazing day to day. Copilot is not even comparable
| there. I like but rarely use aider and plandex. I'd use them
| more if the interface didn't take me completely away from the
| ide. Currently they're closer to "work on this while I'm taking
| a break".
| DeathArrow wrote:
| Have you tried latest Copilot with workspace where you can
| use Claude and add files to the context?
| KronisLV wrote:
| GitHub Copilot in either VS Code or JetBrains IDEs. Having more
| or less the same experience across multiple tools is lovely and
| meets me where I am, instead of making me get a new tool.
|
| The chat is okay, the autocomplete is also really pleasant for
| snippets and anything boilerplate heavy. The context awareness
| also helps. No advanced features like creating entirely new
| structures of files, though.
|
| Of course, I'll probably explore additional tools in the
| future, but for now LLMs are useful in my coding and also
| sometimes help me figure out what I should Google, because
| nowadays seemingly accurate search terms return trash.
| skp1995 wrote:
| yeah I am also getting the sense that people want tooling
| which meets them in their preferred environment.
|
| Do you use any of the AI features which go for editing
| multiple files or doing a lot more in the same instruction?
| adriand wrote:
| I tried Cursor and found it annoying. I don't really like
| talking to AI in IDE chat windows. For whatever reason, I
| really prefer a web browser. I also didn't like the overall
| experience.
|
| I'm still using Copilot in VS Code every day. I recently
| switched from OpenAI to Claude for the browser-based chat stuff
| and I really like it. The UI for coding assistance in Claude is
| excellent. Very well thought out.
|
| Claude also has a nice feature called Projects where you can
| upload a bunch of stuff to build context which is great - so
| for instance if you are doing an API integration you can dump
| all the API docs into the project and then every chat you have
| has that context available.
|
| As with all the AI tools you have to be quite careful. I do
| find that errors slip into my code more easily when I am not
| writing it all myself. Reading (or worse, skimming) source code
| is just different than writing it. However, between type safety
| and unit testing, I find I get rid of the bugs pretty quickly
| and overall my productivity is multiples of what it was before.
| thomasfromcdnjs wrote:
| This is me also, I don't like the UX/DX of Cursor and such
| just yet.
|
| I can't tell if it is a UX thing or if it also doesn't suit
| my mental model.
|
| I religiously use Copilot, and then paste stuff into Claude
| or ChatGPT (both pro) when needed.
| vbezhenar wrote:
| I'm using Copilot in VScode every day, it works fine, but I
| mostly use it as glorified one-line autocomplete. I almost
| never accept multi-line suggestions, don't even look at them.
|
| I tried to use AI deeper, like using aider, but so far I just
| don't like it. I'm very sensitive to the tiny details of code
| and AI almost never got it right. I guess actually the main
| reason that I don't like AI is that I love to write code,
| simple as that. I don't want to automate that part of my work.
| I'm fine with trivial autocompletes, but I'm not fine with
| releasing control over the entire code.
|
| What I would love is to automate interaction with other humans.
| I don't want to talk to colleagues, boss or other people. I
| want AI to do so and present me some short extracts.
| yen223 wrote:
| I am on day 8 of Cursor's 14-day trial. If things continue to
| go well, I will be switching from Webstorm to Cursor for my
| Typescript projects.
|
| The AI integrations are a huge productivity boost. There is a
| substantial difference in the quality of the AI suggestions
| between using Claude on the side, and having Claude be deeply
| integrated in the codebase.
|
| I think I accepted about 60-70% of the suggestions Cursor
| provided.
|
| Some highlights of Cursor:
|
| - Wrote about 80% of a Vite plugin for consolidating articles
| in my blog (built on remix.run)
|
| - Wrote a Github Action for automated deployments. Using Cursor
| to write automation scripts is a tangible productivity boost.
|
| - Made meaningful alterations to a libpg_query fork that
| allowed it to be cross-compiled to iOS. I have very little
| experience with C compilation, it would have taken me a
| substantially long time to figure this out.
|
| There are some downsides to using Cursor though:
|
| - Cursor can get too eager with its suggestions, and I'm not
| seeing any easy way to temporarily or conditionally turn them
| off. This was especially bad when I was writing blog posts.
|
| - Cursor does really well with Bash and Typescript, but does
| not work very well with Kotlin or Swift.
|
| - This is a personal thing, but I'm still not used to some of
| the shortcuts that Cursor uses (Cursor is built on top of
| VSCode).
| skp1995 wrote:
| Its great that cursor is working for you. I do think LLMs in
| general are far far better on Typescript and Python compared
| to other languages (reflects from the training data)
|
| What features of cursor were the most compelling to you? I
| know their autocomplete experience is elite but wondering if
| there are other features which you use often!
| yen223 wrote:
| Their autocomplete experience is decent, but I've gotten
| the most value out of Cursor's "chat + codebase context"
| (no idea what it's called). The feature where you feed it
| the entire codebase as part of the context, and let Cursor
| suggest changes to any parts of the codebase.
| skp1995 wrote:
| ohh inserting.. I tried it on couple of big repos and it
| was a bit of a miss to me. How large are the codebases on
| which you work? I want to get a sense check on where the
| behavior detoriates with embedding + gpt3.5 based
| reranker search (not sure if they are doing more now!)
| yen223 wrote:
| Largest repo I used with Cursor was about 600,000 lines
| long
| skp1995 wrote:
| that's a good metric to aim for... creating a full local
| index for 600k lines is pretty expensive but there are a
| bunch of huristics which can take us pretty far
|
| - looking at git commits - making use of recently
| accesses files - keyword search
|
| If I set these constraints and allow for maybe around 2
| LLM round trips we can get pretty far in terms of
| performance.
| BoorishBears wrote:
| I would not be able to leave a Jetbrains product for Kotlin,
| or XCode for Swift
|
| Overall it's so unfortunate that Jetbrains doesn't have a
| Cursor-level AI plugin* because Jetbrains IDEs by themselves
| are _so much_ more powerful than base level VS Code it
| actually erases some small portion of the gains from AI...
|
| (* people will link many Jetbrains AI plugins, but none are
| polished enough)
| yen223 wrote:
| I probably would switch to Cursor for Swift projects too if
| it weren't for the fact that I will still need Xcode to
| compile the app.
|
| I also agree with the non-AI parts of JetBrains stuff being
| much better than the non-AI parts of Cursor. Jetbrain's
| refactoring tools is still very unmatched.
|
| That said, I think the AI part is compelling enough to
| warrant the switch. There are code rewrite tasks that
| JetBrains would struggle with, that LLMs can do fairly
| easily.
| skp1995 wrote:
| JetBrains is very interesting, what are the best
| performing extensions out there for it?
|
| I do wonder what api level access do we get over there as
| well. For sidecar to run, we need LSP + a web/panel for
| the ux part (deeper editor layer like undo and redo stack
| access will also be cool but not totally necessary)
| mike_hearn wrote:
| You can get both by using Aider (yes, confusingly similar
| name). https://aider.chat
|
| It does the multi-file editing with asking to add files
| etc, but as a CLI/local web app tool.
| jaylane wrote:
| Vscode + cline + openrouter using claude sonnet 3.5 20241022
| model it's unreal the shit it can do
| jeswin wrote:
| I've been building and using these tools for well more than a
| year now, so here's my journey on building and using them
| (ORDER BY DESC datetime).
|
| (1) My view now (Nov 2024) is that code building is very
| conversational and iterative. You need to be able to tweak
| aspects of generated code by talking to the LLM. For example:
| "Can you use a params object instead of individual parameters
| in addToCart?". You also need the ability to sync generated
| code into your project, run it, and pipe any errors back into
| the model for refinement. So basically, a very incremental
| approach to writing it.
|
| For this I made a Chrome plugin, which allowed ChatGPT and
| Claude to edit source code (using Chrome's File System APIs).
| You can see a video here:
| https://www.youtube.com/watch?v=HHzqlI6LLp8
|
| The code is here, but its WIP and for very early users; so
| please don't give negative reviews yet:
| https://github.com/codespin-ai/codespin-chrome-extension
|
| (2) Earlier this year, I thought I should build a VS Code
| plugin. It actually works quite well, allows you to edit code
| without leaving VSCode. It does stuff like adding dependencies,
| model selection, prompt histories, sharing git diffs etc.
| Towards the end, I was convinced that edits need to be
| conversations, and hence I don't use it as much these days.
|
| Link: https://github.com/codespin-ai/codespin-vscode-extension
|
| (3) Prior to that (2023), I built this same thing in CLI. The
| idea was that you'd include prompt files in your project, and
| say something like `my-magical-tool gen prompt.md`. Code would
| be mostly written as markdown prompt files, and almost never
| edited directly. In the end, I felt that some form of IDE
| integration is required - which led to the VSCode extension
| above.
|
| Link: https://github.com/codespin-ai/codespin
|
| All of these tools were primarily built with AI. So these are
| not hypotheticals. In addition, I've built half a dozen
| projects with it; some of it code running in production and
| hobby stuff like webjsx.org.
|
| Basically, my takeaway is this: code editing is conversational.
| You need to design a project to be AI-friendly, which means
| smaller, modular code which can be easily understood by LLMs.
| Also, my way of using AI is not auto-complete based; I prefer
| generating from higher level inputs spanning multiple files.
| skp1995 wrote:
| thats a great way to build a tool which solves your need.
|
| In Aide as well, we realised that the major missing loop was
| the self-correction one, it needs to iteratively expand and
| do more
|
| Our proactive agent is our first stab at that, and we also
| realised that the flow from chat -> edit needs to be very
| free form and the edits are a bit more high level.
|
| I do think you will find value in Aide, do let me know if you
| got a chance to try it out
| jeswin wrote:
| > I do think you will find value in Aide, do let me know if
| you got a chance to try it out
|
| Absolutely, will do it over the weekend. Best of luck with
| the launch.
| DeathArrow wrote:
| >I do think you will find value in Aide, do let me know if
| you got a chance to try it out
|
| Is there a relationship between Aide and Aider or it's just
| a name resemblance?
| skp1995 wrote:
| just a name resemblance, no relationship otherwise
| DeathArrow wrote:
| But why use the web interface instead of Copilot, Cursor,
| Zed, Cline, Aider?
| PittleyDunkin wrote:
| It really depends on what you're doing. AI is great for
| generating a ton of text at once but only a small subset of
| programming tasks clearly benefit from this.
|
| Outside of this it's an autocomplete that's generally 2/3rds
| incorrect. If you keep coding as you normally do and accept
| correct solutions as they appear you'll see a few percentage
| productivity increase.
|
| For highly regular patterns you'll see a drastically improved
| productivity increase. Sadly this is also a small subset of the
| programming space.
|
| One exception might be translating user stories into unit
| tests, but I'm waiting for positive feedback to declare this.
| csomar wrote:
| Neovim + CopilotChat + CopilotLSP + Copilot subscription. You
| can dump your context, autocomplete, chat and select Claude or
| O1. Best deal so far.
| paradite wrote:
| I was deep into AI coding experiments since last December
| before all the VS Code Extensions and IDEs came out.
|
| I wrote a few scripts to get to a semi-automated workflow where
| I have control over the source code context and code editing
| portion, because I believe I can do a better than AI in those
| areas.
|
| Eventually I built my own desktop app which is 16x Prompt:
| https://prompt.16x.engineer/
| vdvsvwvwvwvwv wrote:
| VS Code plugins. Codeium at home. GitHub Copilot at work. Botb
| are good. Probably equivalent.
|
| Codeium recently pushed annoying update that limits your ctrl-I
| prompt to one line and is lost if you lose focus eg to check
| another file. There is a GH issue for that.
| utf_8x wrote:
| I never got the appeal of having the AI directly in your
| editor, I've tried Copilot and whatever JetBrains are calling
| their assistant and I found it mostly just got in the way. So
| for me it's no AI in editor and ChatGPT in a browser for when I
| do need some help.
| davedx wrote:
| I'm still copy/pasting between VS.Code and ChatGPT. I just
| don't want to invest/commit yet because this workflow is _good
| enough_ for me. It lets me chat design, architecture, UX,
| product in the same context as the code which I find helpful.
|
| Pros
|
| - Only one subscription needed
|
| - Very simple
|
| - Highly flexible/adaptive to what part of workflow I'm in
|
| Cons
|
| - More legwork
|
| - Copy/pasting sometimes results in errors due to incomplete
| context
| dagw wrote:
| _what does the AI coding setup of the HN community look like_
|
| GitHub copilot and copilot chat in Jetbrains IDE. Cut and paste
| to Claude for anything else.
| gman83 wrote:
| Why is a fork required? I use the cline plugin for VS Code and it
| seems to be able to be able to more things, like update code
| directly, create new files, etc.
| skp1995 wrote:
| fork was necessary for the UX we wanted to go for. I do agree
| that an extension can also satisfy your needs (and it clearly
| is in your case)
|
| Having a deeper integration with the editor allows for some
| really nice paradigms: - Rollbacks feel more native, in the
| sense that I do not loose my undo or redo stack - cmd+k is more
| in line with what you would expect with a floating widget for
| input instead of it being shown at the very top of your screen
| which is the case with any extension for now.
|
| Going further, the changes which Microsoft are making to enable
| copilot editing features are only open to "copilot-chat" and no
| other extension (fair game for Microsoft IMHO) So keeping these
| things in mind, we designed the architecture in a way that we
| can go towards any interface (editor/extension). We did put
| energy into making this work deeply with the VSCode ecosystem
| of APIs and also added our own.
|
| If the editor does not work to our benefit, we will take a call
| on moving to a different interface and thats where an extension
| or cloud based solution might also make sense
| rco8786 wrote:
| After using Cursor (another AI focused fork) I'm 100% on the
| fork train. AI built natively into the IDE presents another
| layer of speed and isn't subject to the limitations of the
| extension system (which is awesome in its own right, not a
| knock on it).
| jaylane wrote:
| I was on the fork train for awhile but cursors keeps having
| weird issues with indexing, intellisense, not being able to
| save files when format on save is enabled I wound up going
| back to vscode with cline and use openrouter to save money
| via prompt caching. To my knowledge cursor doesn't have
| Claude sonnets computer use enabled yet which is a total game
| changer and cline does I'll check back in a few months but
| instead of paying 20 a month for cursor pro I can put 20 in
| credits in openrouter and fully leverage the latest Claude
| model and features
| skp1995 wrote:
| how do you use the computer usage.. I do find it a very
| interesting API layer to play around with
| DeathArrow wrote:
| But doesn't Cline consume lots of tokens?
| SilentM68 wrote:
| Hmm, any time frame for when Linux (.deb,flatpak) binaries will
| be available?
| skp1995 wrote:
| you should be able to use this:
| https://github.com/codestoryai/binaries/releases/download/1....
| let me know if that does not work.
|
| All our binaries are listed out here:
| https://github.com/codestoryai/binaries/releases/tag/1.94.2....
| ghostwriternr wrote:
| You could also use this script to setup everything: curl -sL
| https://raw.githubusercontent.com/codestoryai/binaries/main/.
| .. | bash (you can see the source of the script too)
| DrBenCarson wrote:
| _sigh_ more Electron
| skp1995 wrote:
| I know... we could have built something free form ground up
| (like zed did) but we had to pick a battle between building a
| new editor from the grounds up or building from a solid
| foundation (VSCode) We are a small team right now (4 of us) and
| have been users of VSCode, so instead of building something
| new, putting energy into building from VSCode made a lot more
| sense to us.
| jamie_ca wrote:
| FYI the youtube embed on https://docs.codestory.ai/features is
| broken (both Firefox and Chrome, MacOS).
|
| https://support.mozilla.org/1/firefox/132.0.1/Darwin/en-US/x...
| skp1995 wrote:
| RIP, didn't expect that to happen. This is the embedded video
| btw https://www.youtube.com/watch?v=i8ZXMgnFSo8 putting it here
| for prosperity
| random42 wrote:
| You mean "posterity", not "prosperity" :)
| CyberCatMeow wrote:
| I see Qwen 2.5 is not listed on your website, is plugging in
| different llms supported as well?
| skp1995 wrote:
| Honestly we can, I haven't prompted it enough what do you want
| to use the model for?
| CyberCatMeow wrote:
| Just general coding, mostly python. Seems to me that Qwen
| 2.5, especially the upcoming bigger coder model might be the
| best performing coding model for 24GB VRAM setups
| jwilber wrote:
| This looks great. Would love some blog posts about your
| experience building this our with rust!
| skp1995 wrote:
| Oh for sure! I do want to talk about how rust really helped us
| so many times when doing refactors or building new features,
| part of the reason why we were able to iterate so quickly on
| the AI side of things and ship features
| ilrwbwrkhv wrote:
| This is very similar to the Zed editor. How much did you get
| inspired by them? And what are the differences between yours and
| their implementations?
| skp1995 wrote:
| I would take that as a compliment, big fan of Zed (I hope their
| extension ecosystem allows for us to plugin sidecar into Zed
| soon)
|
| Tbh I did try out their implementation and it still feels
| early, one of the key difference we went for was to allow the
| user to move freely between chat and editing mode.
| ilrwbwrkhv wrote:
| There's feels much more detailed and thoughtful than yours.
|
| Yours for example doesn't allow one to insert diagnostics or
| allow for example for things like inserting all tabs open and
| all files open at once.
|
| They also allow jumping from editing to chatting mode by
| simply doing command enter.
| skp1995 wrote:
| hmmm you are right, the ergonomics of providing context are
| more powerful in zed. Feedback taken, we will work on it.
|
| We implicitly take in all the diagnostics on the files http
| s://github.com/codestoryai/sidecar/blob/e5408782a3bfa461...
| rco8786 wrote:
| Is there a comparison with Cursor I can read?
| nsonha wrote:
| first editor I've seen recently that defaults to turn of minimap.
|
| I won't shut up for about this, I don't understand how such an
| useless "feature" becomes the norm in modern IDEs.
| gitgud wrote:
| This is a fork of VScode, which means people can't use the
| extension store anymore right?
| skp1995 wrote:
| They can from the openvsx store https://open-vsx.org/
|
| We also import your extensions automatically (safe guarding
| against the ones with Microsoft's licensed)
|
| You can also just download in from the vscode marketplace
| webpage and drag and drop it in
| bagels wrote:
| What is the privacy policy? Do you get to see my code and
| secrets? Does anybody else? I don't want you to. Nothing
| personal.
| skp1995 wrote:
| we don't .. in your user settings, type in disable all
| telemetry and we won't see a thing
| edf13 wrote:
| What telemetry are you sending without this setting?
| skp1995 wrote:
| nothing https://github.com/codestoryai/ide/blob/0eb311b7e4d
| 7d63676ad... we get back an undefined PosthogClient instead
| of populating it
| PittleyDunkin wrote:
| In a world that prioritizes precise user control over the output
| text, how do you justify the value of relinquishing such control
| even to provide actions the user would have made already? It only
| takes a single bad edit to make the user lose all trust.
| skp1995 wrote:
| yeah bad edits are especially worse when we keep building on
| top of it.
|
| Our first take on solving this is with rollbacks .. which
| allows you to delete edits up until a point in the
| conversation.. so if you do notice a bad edit you can do that.
|
| after this, there is the proactive agent ..which checks it's
| work again and suggests more changes which it needs to do ..
| you can give feedback and guide it.
|
| With llms we do loose a bit of control but I think the editor
| should work to solve this
| PittleyDunkin wrote:
| I appreciate the earnest replies; these do in fact address my
| concerns.
| sagarpatil wrote:
| Been using Cursor since launch. Really frustrating how they
| charge per message (500/mo) instead of by token usage. Like, why
| should a one-line code suggestion cost the same as refactoring an
| entire class? Plus it's been losing context lately and making
| rookie mistakes.
|
| Tried Zed AI but couldn't get into it - just doesn't feel as
| smooth as Cursor used to. GitHub Copilot is even worse. Feels
| like they're so obsessed with keeping that $10/month price point
| that they've gutted everything. Context window is tiny and the
| responses are barebones.
| DeathArrow wrote:
| You can try Cline and Aider.
| tekacs wrote:
| I used Cursor for many months, but found that I couldn't deal
| with how slow and workflow-interrupting VSCode feels, so I went
| back to Zed.
|
| I tried out and abandoned Zed AI, but I've found that Zed +
| aider is a really excellent setup - for me, at least.
|
| For smaller things, Zed's inline Copilot setup works (nowadays)
| just as well as Cursor's and for things that are even a little
| bigger than tiny, I pull up aider and prompt the change the
| exact same way that I did with Cursor's Composer before.
|
| I'm an odd choice because I'm a pretty expert-level programmer,
| but I find that aider with o1 is helpful for things hairy,
| expansive and tedious things that would otherwise irritate.
| nobody9999 wrote:
| Just an FYI, there is extant code called Aide[0][1] (Advanced
| Intrusion Detection Environment) as well that's under active
| development/maintenance.
|
| Since they perform vastly different functions (the above is a
| replacement for tripwire[2]), it's unlikely anyone will be
| confused, but be aware that it exists. There's always someone who
| will mistake one for the other.
|
| [0] https://aide.github.io/
|
| [1]
| https://en.wikipedia.org/wiki/Advanced_Intrusion_Detection_E...
|
| [2] https://en.wikipedia.org/wiki/Open_Source_Tripwire
| DeathArrow wrote:
| If it's a fork of VS code it should be trivial to also support
| Linux and Windows. Why is it MacOs only?
| skp1995 wrote:
| It supports all platforms, we took inspiration from the macos
| spotlight search widget for inline editing.
| santiagobasulto wrote:
| I think it's confusing for people that only one big green
| button shows "Download for mac" and the other download links
| are at the footer.
| skp1995 wrote:
| ugh.. in which case our platform detection code is not
| working as expected. We will look into that
| santiagobasulto wrote:
| There are download links for other platforms at the footer, and
| below the main CTA button says "Looking for other platforms?"
| which takes you there.
| jadbox wrote:
| I'd much, much prefer Aide to continue as a CLI tool or as a
| VSCode plugin. Every fork of VSCode ends up with IDE maintenance
| bugs that never get addressed and slowly the effort implodes as
| the bug surface becomes too wide.
|
| Do you want to spend 90% of your time on AI or troubleshooting
| odd Linux VSCode bugs in your fork? I'd highly recommend the team
| to evaluate a different direction for growth to maximize
| sustainable future growth.
| skp1995 wrote:
| Thats a fair point, a significant part of our 4 person team had
| to skill up on the VSCode codebase to be able to meaningfully
| make changes to it.
|
| I would love to know your workflow, you mention CLI tool or
| VSCode plugin, which one of them work for you? Whats missing
| from them where Aide can fill in the gap
| portpecos wrote:
| Plugin, which gives you access to the UI.
| wg0 wrote:
| Is this "VC less" AI IDE?
| nullspace wrote:
| I have just downloaded and set it up to talk to Anthropic. I have
| some unused API credits that I've been trying to find a use for.
| Haha.
|
| To be honest, I was very confused by the the @'ing. Is it
| possible to always pin the current open file to the context,
| perhaps in a FIFO fashion? (Other than the manually pinned
| files). The UX around pinning / @'ing is not super clear to me.
|
| Otherwise still getting the hang of it, and generally pretty
| neat.
| skp1995 wrote:
| There is pinned context on the very top where you can pin the
| files which you frequently use.
|
| We will start including the open file by default in the context
| very soon (one of the gotchas here, is that the open file could
| not be related to the question you have)
___________________________________________________________________
(page generated 2024-11-07 23:01 UTC)