Love the idea of this, and very excited to see how it pans out. That said: I hate the code review UI. Just dump the changes as `git diff` does and let me review them using all the code review tools I use every day, then provide revision instructions. Building a high-quality TUI for side-by-side diffs should not be the thing you are spending time on, and there already exist great tools for viewing diffs in the termina…
Thanks for the feedback! I actually had a ‘plandex diff’ command working at one point, but dropped it in favor of the changes TUI. I could definitely bring it back for people who prefer that format.
Show HN: Plandex – an AI coding engine for complex tasks
101–110 of 112 posts
Re: Show HN: Plandex – an AI coding engine for complex tasks
#102Earlier quoted context omitted.
Thanks for the feedback! I actually had a ‘plandex diff’ command working at one point, but dropped it in favor of the changes TUI. I could definitely bring it back for people who prefer that format.
Providing diff output allows people to self select their approach to merging the changes.
Re: Show HN: Plandex – an AI coding engine for complex tasks
#103Earlier quoted context omitted.
Thanks, I'm aware of ollama and the open source model ecosystem, but I haven't done a deep dive yet, so all the info in this thread has been quite helpful.
In theory, all you have to do is redirect the API gateway to localhost and all your existing integrations should just work!
It seems that while ollama does have partial OpenAI API compatibility, it's missing function calling, so that's a blocker for now.
Re: Show HN: Plandex – an AI coding engine for complex tasks
#104Re: Show HN: Plandex – an AI coding engine for complex tasks
#105Re: Show HN: Plandex – an AI coding engine for complex tasks
#106Whats the deal with plandex cloud and $10/$20-mo? The github repo README devolves into a cloud pitch halfway through. I thought this was a local binary talking to openAI? I thought this was open source?
The CLI talks to the Plandex server and the server talks to OpenAI.
Re: Show HN: Plandex – an AI coding engine for complex tasks
#107Very small nit: it'd be nice to provide an OpenAI org in case multiple orgs exist.
Re: Show HN: Plandex – an AI coding engine for complex tasks
#108Whats the deal with plandex cloud and $10/$20-mo? The github repo README devolves into a cloud pitch halfway through. I thought this was a local binary talking to openAI? I thought this was open source?
Hi, it’s open source and it also has a cloud option. You can either self-host or use cloud—it’s up to you. The CLI talks to the Plandex server and the server talks to OpenAI.
Re: Show HN: Plandex – an AI coding engine for complex tasks
#109Earlier quoted context omitted.
Hi, it’s open source and it also has a cloud option. You can either self-host or use cloud—it’s up to you. The CLI talks to the Plandex server and the server talks to OpenAI.
but i still don't get what the cloud option would be doing that's worth $20/mo if it's talking to openAI. Does the plandex server have large resource requirements?
I actually did start out with just the CLI running locally, but it reached a point I needed a database and thus a client-server model to get it all working smoothly. I also want to add sharing and collaboration features in the future, and those require a client-server model.
Re: Show HN: Plandex – an AI coding engine for complex tasks
#110Earlier quoted context omitted.
I agree, these things seem to do okish on trivial web projects. I've never seen them do anything more than that. I still use ChatGPT for some coding tasks, e.g. I asked it to write C code to do some annoying fork/execve stuff (can't remember the details) and it did a decentish job, but it's like 90% right. Great for figuring out a rough shape and what functions to search for, but you definitely can't just take the co…
I used a web project in the demo because I figured it would be familiar to a wide range of developers, but actually many nontrivial pieces of Plandex have been built with the help of Plandex itself. That's not to say it's perfect or will never make "noob-level" mistakes. That can definitely happen and is ultimately a function of the underlying model's intelligence. But I can at least assure you that it's quite capabl…