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

A desktop app for isolated, parallel agentic development

github.com

Post reply on HN
Print view View original

21–30 of 51 posts

Re: A desktop app for isolated, parallel agentic development

#21

Post by lucid-dev »

ttoinou wrote: post #2

Something I never see mention is that all agents / CLI tools seem to modify local files. Which makes editing current files, working with the git, asking to only modify some parts of the files etc. cumbersome, as the agent is constantly reading and writing to files that we are also accessing. This is usually solved by using git worktree but this solution requires 1 new folder name and branch for each new agent, and ea…

You use git worktrees, and then merge-in. Or rebase, or 3-way merge, as necessary.

I have a local application I developed that works extremely well for this. I.e. every thread tied to a repo creates it's own worktree, then makes it edits locally, and then I sync back to main. When conflicts occur, they are either resolved automatically if possible (i.e. another worktree merged into main first, those changes are kept so long as they don't conflict, if conflicted we get the opportunity to resolve, etc.).

At any merge-into-main from a worktree, the "non-touched" files in the worktree are automatically re-synced to main, thus updating the worktree with any other changes from any other worktree that have been already pushed to main.

Of course, multiple branches can also be used and then eventually merged into a single branch later..

---

Also, this is very clearly exactly the same thing OP does in their system, as per the README on their github link..

Re: A desktop app for isolated, parallel agentic development

#22

Post by CGamesPlay »

ttoinou wrote: post #2

Something I never see mention is that all agents / CLI tools seem to modify local files. Which makes editing current files, working with the git, asking to only modify some parts of the files etc. cumbersome, as the agent is constantly reading and writing to files that we are also accessing. This is usually solved by using git worktree but this solution requires 1 new folder name and branch for each new agent, and ea…

> While it can be super powerful, I wish there was a quicker "in memory" agent solution where each agent keeps in its own RAM the list of files modifications ("patch") it recommends to apply to solve current issue. Then we could apply that patch depending on what we're doing, if we have others patches to apply before etc.

Then the changes can't be tested to even verify that they pass the compiler/linter, much less tested to confirm they actually work. The only way to fix this is to "modify local files", where "local files" is either a separate worktree (that you don't manage or need to know the location of, possibly even on a separate machine); or a hacky, vibe-coded, in-memory VFS; or somewhere in between.

Re: A desktop app for isolated, parallel agentic development

#23

Post by plutonium3345 »

ttoinou wrote: post #2

Something I never see mention is that all agents / CLI tools seem to modify local files. Which makes editing current files, working with the git, asking to only modify some parts of the files etc. cumbersome, as the agent is constantly reading and writing to files that we are also accessing. This is usually solved by using git worktree but this solution requires 1 new folder name and branch for each new agent, and ea…

I personally have had a frustrating experience with GitButler, and would prefer if things don't become too complicated. For example, when you accidentally break GitButler, it can become really difficult to recover all your unpushed progress. It is also hard to find the exact location of where your code changes are stored across different branches.

My suggestion would be to keep things simple, pragmatic, and save development time. While git worktrees are not perfect, and they require extra space, it is easier for people to understand and easily locate this kind of structure while also being able to execute commands in a somewhat isolated environment. I would be happy if the app simply automates this for me and creates a new worktree, branch, and agent at that location as soon as I click 'Add agent'.

Then the only issues become merge conflicts between different branches... this is where extra dev time could be allocated, and implement agents that automatically merge branches.

Re: A desktop app for isolated, parallel agentic development

#24

Post by plutonium3345 »

IanCal wrote: post #3
ttoinou wrote: post #2

Something I never see mention is that all agents / CLI tools seem to modify local files. Which makes editing current files, working with the git, asking to only modify some parts of the files etc. cumbersome, as the agent is constantly reading and writing to files that we are also accessing. This is usually solved by using git worktree but this solution requires 1 new folder name and branch for each new agent, and ea…

This is a good point. Docker? I typically want other kinds of isolation for services and things anyway - it’s got its own file system, you could have N versions on the same branch working without conflict (at least the conflict you’d have with work trees). There is some more plumbing involved but…not much? Edit - task boards are my first thought for the comms side for agents sharing info.

Yeah, I think as agents become more capable, more isolation will be necessary. Hence, I also agree that either containers or VMs will eventually be required. We can see how tools like Cursor already have a built-in browser so that the agent can "see" (probably as text for now) what component you want to modify. In the future, I believe the workflow for an agent will be something like: [make changes] -> [get user input] -> [take a screenshot] -> [process the screenshot and user input] -> [make changes].

I doubt something like this can be implemented easily in a single environment without running into client and server port issues, etc.

Re: A desktop app for isolated, parallel agentic development

#25

Post by alganet »

ttoinou wrote: post #2

Something I never see mention is that all agents / CLI tools seem to modify local files. Which makes editing current files, working with the git, asking to only modify some parts of the files etc. cumbersome, as the agent is constantly reading and writing to files that we are also accessing. This is usually solved by using git worktree but this solution requires 1 new folder name and branch for each new agent, and ea…

It is very likely that agent tooling will get better at doing asynchronous things and being aware of the user interacting in parallel, without git, and probably within a single session.

It's feasible, and it makes more sense than separating and merging later (or keeping patches in memory then applying in bulk).

Why do I say it's feasible? We have the technology, right? The IDE knows which file the user has in focus, and can orient agents to use tooling that would inform them of that fact when they're running. Similarly, that same tooling could just spend a little bit of time planning focus to spread to multiple agents in a way they won't overlap.

Maybe big repos, monorepos and so on are a limitation. If we were on the previous "small-to-middle interlinked projects" era, that division would come in naturally. You only really need multiple agents in parallel on a single project if that thing is big enough to have more than one angle to work on. It's a push-and-pull that changes with the times, maybe we're heading to a more granular way of doing things.

Re: A desktop app for isolated, parallel agentic development

#26

Post by ttobi »

nojs wrote: post #7

This looks very useful. I would love something similar that lets me plug in actual Claude Code/Codex with their original agent loop, prompting etc, and just handles the multiplexing, worktrees, isolation, etc automatically (it looks like this tool doesn’t support that). Because I think a lot of the power of eg CC comes from the engineering they’ve done to the tool rather than the underlying model. How are people doin…

I have build my own tool for this https://github.com/tobias-walle/agency It uses tmux to run the agents and offers some convenience commands that, e.g. lets you merge the changes back into its original branch. I added some simple idle and change detection, so you can see which agent needs your attention. As all agents are just simple cli commands it is very easy to extend the config with your tool of choice.

Re: A desktop app for isolated, parallel agentic development

#27

Post by justinmayer »

Interesting to see two similar projects with the same "cmux" name, at least until this one was renamed two weeks ago to "mux". Naming is hard.

I have not yet tried either one, but here is the other project for those who want to compare and contrast them:

https://github.com/manaflow-ai/cmux

Re: A desktop app for isolated, parallel agentic development

#28

Post by undeveloper »

ttoinou wrote: post #2

Something I never see mention is that all agents / CLI tools seem to modify local files. Which makes editing current files, working with the git, asking to only modify some parts of the files etc. cumbersome, as the agent is constantly reading and writing to files that we are also accessing. This is usually solved by using git worktree but this solution requires 1 new folder name and branch for each new agent, and ea…

gitbutler is neat, but doesn't separate out files -- the changes are still visible to other sessions

Re: A desktop app for isolated, parallel agentic development

#29

Post by brainless »

ttoinou wrote: post #2

Something I never see mention is that all agents / CLI tools seem to modify local files. Which makes editing current files, working with the git, asking to only modify some parts of the files etc. cumbersome, as the agent is constantly reading and writing to files that we are also accessing. This is usually solved by using git worktree but this solution requires 1 new folder name and branch for each new agent, and ea…

In my coding agent, nocodo (1), I am thinking about using copy on write filesystems for cheaper multi-agent operations. But to be honest git worktree may be good enough for most use cases. nocodo checks existing worktree in the local repo and I will add creation and merge support too.

1. https://github.com/brainless/nocodo

Re: A desktop app for isolated, parallel agentic development

#30

Post by videlov »

ttoinou wrote: post #2

Something I never see mention is that all agents / CLI tools seem to modify local files. Which makes editing current files, working with the git, asking to only modify some parts of the files etc. cumbersome, as the agent is constantly reading and writing to files that we are also accessing. This is usually solved by using git worktree but this solution requires 1 new folder name and branch for each new agent, and ea…

(co-founder of GitButler here)

We chose not to use separate git worktrees under the hood for this functionality. Let me try to break down why, maybe there's an opportunity for me to learn more here.

In my head I separate between use cases of 1) "different tasks" and 2) "best of n, same task".

The app that we built already had the ability to separate changes into branches while in the worktree (on disk) it renders the integration of the branches. Our canonical use case back in the days was "A developer works on a feature branch and wishes to commit & publish a bugfix from separate branch". When we learned that people were using this for running multiple parallel agents we added some additional tooling for it.

So in practice what happens when you have multiple agents coding in parallel with GitButler is that the system captures information after an agent completes an edit (via the agent hooks) and uses that to 1) stage the particular edit to a branch dedicated to the agent and 2) perform a commit into that branch (GB can have multiple staging areas, one per applied branch).

The system will not allow multiple agents to edit the same file at the same time (via a locking mechanism in the pre-edit hook), but agents do see each others changes.

In the context of the "different tasks for different agents" use case, we have found that them seeing edits by others to have a positive effect on the outcomes. The first one that comes to mind is - no merge conflicts. But beyond merge conflicts, we have found that there is a lower likelihood of reaching a state where code diverges semantically.

In my own usage, I have found it helpful when I am hands on programming on something and wish to have an agent do some auxiliary task, for us to share a workspace (so that I can nudge it one way or another).

Is there something I am missing here? Of course for best-of-n of the same task this doesn't exactly make sense, but with regards to different tasks, what are some additional reasons to require full isolation? (as different worktrees would provide)

Post reply on HN