To me this (coupled with their DeltaDB announcement[0]) feels like Zed trying to get out of the business that Cursor is in, and maybe even kicking them on the way out, if the protocol helps bring-your-own-agent take off. These agent tools are already not particularly sticky, especially when they are outside the editor. I don't use Cursor or Zed full time, but I have heard that the most compelling bit is the tab completion, which isn't even part of the agent stuff. I'm sure it's nice, but it's very hard to imagine that being a moat for an editor.
I think I agree with the implicit judgment from the Zed team that it's too early to try to lock people in and capture value, and the best way to build long-term user loyalty is to just be the best editor and let people work however they want to work. On top of that, it is not a great use of their dev time to iterate on the details of an agentic coding tool when there are 10 other ones doing the same thing, and any prompting secret sauce a) is trivial to copy, and b) gets eaten by the next model generation anyway.
I don't really understand what Cursor's business model is supposed to be long term -- at least Zed is trying to come up with new things that an editor can be (see their chat ideas and now source control). On the other hand, I also don't buy the argument some have made that Cursor and friends were banking on the marginal cost of inference going to zero, keeping prices at $20 a month, and pocketing the difference. It seems obvious (even a year ago, but moreso now) that in that situation, the IDEs also compete down to zero. If anything, higher total inference spend has to be better for them: more to skim off of. If you're already spending $50 a month on LLM tooling, Cursor doesn't have to be that big of a value-add to get you to pay them $52 instead.
I purchased Panic's Nova editor when it was first released. I'd be willing to pay Zed an equivalent ~79 per year for a perpetual license (with updates after the first year requiring further payment).
Obviously the protocol is designed to encourage other agents to build their own integration. It seems like the kind of thing LLMs would be good at, so I expect to see it soon, especially for the open source ones.
I like Zed. It's super fast and seems to be getting lots of improvement constantly.
My issue with it is when I use it on my codebase, it doesn't like my (probably old style) eslintrc. So it decides to go ahead and reformat my file on save :(.
For anyone using Zed, how extensive is its ability to be configured/modded to emulate other editor behaviors, bindings, etc? Eg i want something akin to Helix, but i'm not interested in Zeds Vim mode. Nor am i interested in a straight Helix mode, as my own Helix usage is quite customized.
Can i make my own Modes trees like Helix/etc offers? I'm aware of https://zed.dev/docs/helix, just not sure what the UX is to customize this behavior yourself as a user.
not to be confused with the other ACP from IBM, Agent Communication Protocol... Which is about "communication between agents, humans, and applications".
Threads for ongoing work, runs for stateless one-shots, long term memory. And a discovery system. No affiliation, just fun to see what folks have in their APIs.
It's very 'batteries-included', for one thing - when a novice wants to code, I recommend them Zed because it'll just handle and manage LSPs for them for a variety of languages. Meanwhile with VSCode step 1 of installing and using it for e.g. Rust is to go and install a random extension (and the VSCode store, whilst sorted by popularity, can be intimidating / confusing for a novice, who might install a random/scammy extension). The 'recommended extensions' thing helps, but it's still subpar.
It has some other niceties – I love how if you Cmd+Shift+F to search across the project, that you get a multi-buffer [1] - I often use that for larger manual refactors for a ton of places in my codebase.
But honestly... as others have said, speed is just _such_ a strong feature for my taste - it makes a world of difference compared to VSCode, because in VSC I'll be typing vim commands, the editor will fail to keep up and it'll do the wrong thing - whereas in Zed it's fast enough that I never really run into stalls.
Having everything be an extension is the double edged sword of VS Code. Zed is great for the ecosystem and I use it as an alternate editor for quick text editing but I dont foresee it replacing VS Code as my IDE. Once youve configured VS Code to your liking with devcontainers, and extensions declared by the config file, it becomes excellent.
The biggest problem with VSC for me is that sometimes undo history is completely broken with VIM. If you don't commit frequently, it is very easy to mess up the with the project and lose all your work, if you undo anything.
When I heared that Zed is VC backed, it was a signal to move away from it. Emacs, NeoVim will be here in fifty years from now. VC backed companies don’t have a guarantee. VCs except too much, and pressure founders to pivot if they don’t seem to fulfill the promise for 20x return. See what happened to Evernote. Or SoundCloud.
That doesn't guarantee it will have paid contributors indefinitely, but the same is true of the other editors you listed. It does, however, guarantee that if Zed (the company) were to disappear, community members would be free to continue using the Zed editor (and developing it) in perpetuity!
The problem is the concern of future development. Zed the company controls what gets merged. If it is profitable to leave some features out, will they leave them even if there is PR? Or something else. Compare it to the problems of Chromium.
My guess is that there aren't that many folks using zed over neovim/emacs. It seems like a more suitable alternative to vscode or jetbrains IDEs which imo have at least the same level of financial risks involved. At the end of the day it is just an editor and moving to something is pretty minimal effort if things do go in a direction you don't agree with.
I get enshitification in general, but your take is absurd. zed is merely a text editor, a tool operate on your local files with no special format, no lock in. It may charge you $1m a year , or show you one ad every 5 sec, then you don't have to use it...
To me this (coupled with their DeltaDB announcement[0]) feels like Zed trying to get out of the business that Cursor is in, and maybe even kicking them on the way out, if the protocol helps bring-your-own-agent take off. These agent tools are already not particularly sticky, especially when they are outside the editor. I don't use Cursor or Zed full time, but I have heard that the most compelling bit is the tab completion, which isn't even part of the agent stuff. I'm sure it's nice, but it's very hard to imagine that being a moat for an editor.
I think I agree with the implicit judgment from the Zed team that it's too early to try to lock people in and capture value, and the best way to build long-term user loyalty is to just be the best editor and let people work however they want to work. On top of that, it is not a great use of their dev time to iterate on the details of an agentic coding tool when there are 10 other ones doing the same thing, and any prompting secret sauce a) is trivial to copy, and b) gets eaten by the next model generation anyway.
I don't really understand what Cursor's business model is supposed to be long term -- at least Zed is trying to come up with new things that an editor can be (see their chat ideas and now source control). On the other hand, I also don't buy the argument some have made that Cursor and friends were banking on the marginal cost of inference going to zero, keeping prices at $20 a month, and pocketing the difference. It seems obvious (even a year ago, but moreso now) that in that situation, the IDEs also compete down to zero. If anything, higher total inference spend has to be better for them: more to skim off of. If you're already spending $50 a month on LLM tooling, Cursor doesn't have to be that big of a value-add to get you to pay them $52 instead.
[0]: https://zed.dev/blog/sequoia-backs-zed#introducing-deltadb-o...
Zed is great. I'd happily pay for it.
I purchased Panic's Nova editor when it was first released. I'd be willing to pay Zed an equivalent ~79 per year for a perpetual license (with updates after the first year requiring further payment).
Okay, but what about Claude Code, the coding agent people actually use?
Obviously the protocol is designed to encourage other agents to build their own integration. It seems like the kind of thing LLMs would be good at, so I expect to see it soon, especially for the open source ones.
I like Zed. It's super fast and seems to be getting lots of improvement constantly.
My issue with it is when I use it on my codebase, it doesn't like my (probably old style) eslintrc. So it decides to go ahead and reformat my file on save :(.
You can disable this in your global or project settings, and even per language: https://zed.dev/docs/configuring-zed#format-on-save
I think this would be pretty trivial to fix zed’s autoformatter settings.
For anyone using Zed, how extensive is its ability to be configured/modded to emulate other editor behaviors, bindings, etc? Eg i want something akin to Helix, but i'm not interested in Zeds Vim mode. Nor am i interested in a straight Helix mode, as my own Helix usage is quite customized.
Can i make my own Modes trees like Helix/etc offers? I'm aware of https://zed.dev/docs/helix, just not sure what the UX is to customize this behavior yourself as a user.
So this ACP is Agent Client Protocol
https://github.com/zed-industries/agent-client-protocol
not to be confused with the other ACP from IBM, Agent Communication Protocol... Which is about "communication between agents, humans, and applications".
https://github.com/i-am-bee/ACP
Which is now part of A2A
https://github.com/a2aproject/A2A
I'm curious if there was an attempt to extend this to add first class support for interaction with IDEs rather than creating a new protocol.
Not to be confused with IBM's ACP: Agent Communication Protocol. C here being Client. https://agentcommunicationprotocol.dev/introduction/welcome
Implemented by langchain, https://github.com/langchain-ai/agent-protocol
Threads for ongoing work, runs for stateless one-shots, long term memory. And a discovery system. No affiliation, just fun to see what folks have in their APIs.
This in theory means you can use QwenCoder too since it’s a fork of Gemini CLI?
as a policy, im not allowing any agentic CLI installed on my machine outside of a container. How can I use Zed with Gemini CLI?
You can configure an arbitrary command to connect to ACP in your settings, maybe something like:
{"agent_servers": { "Container Gemini": { "command": "ssh", "arguments": ["hostname", "gemini", "--experimental-acp"] } }
could work?
similar to https://eca.dev/ ?
Interesting, thanks for linking that. ACP seems simpler. Nice comparison here by gpt-5: https://gist.github.com/david-crespo/e9a9ab88df73d4c2702ef04...
I hope that cursor-agent get support to ACP
which other major agents are on board with Agent Client Protocol?
Related post from Google: https://developers.googleblog.com/pt-br/gemini-cli-is-now-in...
We'll put that link in the top text too. Thanks!
The original link was to Portuguese. This is English version https://developers.googleblog.com/en/gemini-cli-is-now-integ...
Thanks! Replaced.
Honest question - why use Zed, other than it is fast and “not VS Code”?
It's very 'batteries-included', for one thing - when a novice wants to code, I recommend them Zed because it'll just handle and manage LSPs for them for a variety of languages. Meanwhile with VSCode step 1 of installing and using it for e.g. Rust is to go and install a random extension (and the VSCode store, whilst sorted by popularity, can be intimidating / confusing for a novice, who might install a random/scammy extension). The 'recommended extensions' thing helps, but it's still subpar.
It has some other niceties – I love how if you Cmd+Shift+F to search across the project, that you get a multi-buffer [1] - I often use that for larger manual refactors for a ton of places in my codebase.
But honestly... as others have said, speed is just _such_ a strong feature for my taste - it makes a world of difference compared to VSCode, because in VSC I'll be typing vim commands, the editor will fail to keep up and it'll do the wrong thing - whereas in Zed it's fast enough that I never really run into stalls.
[1]: https://zed.dev/features#multi-buffers
Having everything be an extension is the double edged sword of VS Code. Zed is great for the ecosystem and I use it as an alternate editor for quick text editing but I dont foresee it replacing VS Code as my IDE. Once youve configured VS Code to your liking with devcontainers, and extensions declared by the config file, it becomes excellent.
The biggest problem with VSC for me is that sometimes undo history is completely broken with VIM. If you don't commit frequently, it is very easy to mess up the with the project and lose all your work, if you undo anything.
IME, it's not as polished as VSCode, too many rough little details here and there. But then, VSCode has been around for longer.
The vim emulation is very good.
Seconding this, it's the only editor other than emacs that has a vim-mode that feels like it's maintained by a long term vim user.
I use it as a notepad mostly, just because vscode is for the actual repo and sublime would nag me with the trial/purchase thing.
It might sound silly but it helps me to have one use per app (even better if each is clearly identifiable by their theme).
Also very active development, which is pretty notable when every other IDE-lite but Nova and JetBrains seems to have just given up.
It's not Microsoft, which is kind of a big deal because they own so much of developer tool ecosystem right now
When I heared that Zed is VC backed, it was a signal to move away from it. Emacs, NeoVim will be here in fifty years from now. VC backed companies don’t have a guarantee. VCs except too much, and pressure founders to pivot if they don’t seem to fulfill the promise for 20x return. See what happened to Evernote. Or SoundCloud.
Zed is open-source under GPLv3 - https://github.com/zed-industries/zed
That doesn't guarantee it will have paid contributors indefinitely, but the same is true of the other editors you listed. It does, however, guarantee that if Zed (the company) were to disappear, community members would be free to continue using the Zed editor (and developing it) in perpetuity!
The problem is the concern of future development. Zed the company controls what gets merged. If it is profitable to leave some features out, will they leave them even if there is PR? Or something else. Compare it to the problems of Chromium.
Jacob Kaplan-Moss influenced my own thinking around this: https://social.jacobian.org/@jacob/111914179201102152
“We believe that open source should be sustainable and open source maintainers should get paid!”
Maintainer: introduces commercial features
“Not like that”
Maintainer: works for a large tech co
“Not like that”
Maintainer: takes investment
“Not like that”
The company is VC backed. Why would the software disappear though? It’s open source, so it can survive even if the company doesn’t
My guess is that there aren't that many folks using zed over neovim/emacs. It seems like a more suitable alternative to vscode or jetbrains IDEs which imo have at least the same level of financial risks involved. At the end of the day it is just an editor and moving to something is pretty minimal effort if things do go in a direction you don't agree with.
I get enshitification in general, but your take is absurd. zed is merely a text editor, a tool operate on your local files with no special format, no lock in. It may charge you $1m a year , or show you one ad every 5 sec, then you don't have to use it...