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

Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

github.com

Post reply on HN
Print view View original

71–80 of 85 posts

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#71

Post by throwup238 »

danenania wrote: post #70
throwup238 wrote: post #69

Earlier quoted context omitted.

I would get rid of the email validation code for localhost, though. That remains the biggest annoyance when running it locally as a single user. I would also add a $@ to the docker-compose call in the bash start script so users can start it in detached mode.

It should already be skipping the email validation step in local mode. Is it showing up for you? I’ll look into the detached mode, thanks!

Yes, it showed up for me, luckily I had the logs open and remembered that was the solution in v1 (it wasn’t documented back then iirc). I git pulled in the same directory I ran v1 in so maybe there’s some sort of left over config or something?

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#72

Post by killerstorm »

danenania wrote: post #66
killerstorm wrote: post #59

I like the idea but it did not quite work out of box. There was some issue with sign-in, it seems pin requested via web does not work in console (so the web suggesting using --pin option is misleading). I tried BYO plan as I already have openrouter API key. But it seems like default model pack splits its API use between openrouter and openai, and I ended up stuck with "o3-mini does not exist". And my whole motivation…

Thanks for trying it! Could you explain in a bit more detail what went wrong for you with sign-in and the pin? Did you get an error message? On OpenRouter vs. OpenAI, see my other comment in this thread ( https://news.ycombinator.com/item?id=43719681 ). I'll try to make this smoother. On Gemini 2.5 Pro: the new paid 2.5 pro preview will be added soon, which will address this. The free OpenRouter 2.5 pro experimental…

When I used `--pin` argument I got an error message along the lines of "not found in the table".

I got it working by switching to oss model pack and specifying G2.5P on top. Also works with anthropic pack.

But I'm quite disappointed with UX - there's a lot of configuration options but robustness is severely lacking.

Oddly, in the default mode out of box it does not want to discuss the plan with me but just jumps to implementation.

And when it's done writing code it aggressively wants me to decide whether to apply -- there's no option to discuss changes, rewind back to planning, etc. Just "APPLY OR REJECT!!!". Even Ctrl-C does not work! Not what I expected from software focused on planning...

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#73

Post by killerstorm »

vunderba wrote: post #60
killerstorm wrote: post #59

I like the idea but it did not quite work out of box. There was some issue with sign-in, it seems pin requested via web does not work in console (so the web suggesting using --pin option is misleading). I tried BYO plan as I already have openrouter API key. But it seems like default model pack splits its API use between openrouter and openai, and I ended up stuck with "o3-mini does not exist". And my whole motivation…

Yeah, I noticed that (needing a dedicated OpenAI key) as well for the BYO key plan. It's a little bit odd considering that open router has access to the open AI models. https://openrouter.ai/openai

FWIW I got it working by selecting oss or anthropic model packs. But I had some OpenAI key... maybe it would work with a dummy.

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#74

Post by dr_kiszonka »

danenania wrote: post #41
jtwaleson wrote: post #22

Nice! I tried it out when you launched last year but found it pretty expensive to use. I believe I spent $5 for half an hour of coding or so. Can you share what the typical costs are now, since the model prices have changed significantly?

It's a bit hard to give "typical" costs because it's so dependent on how you use it. The project map size (which scales with overall project size) and the number/size of relevant files are the main drivers of cost, so working in large existing codebases will be a lot more expensive than generating a new app from scratch. Taking Plandex's codebase as an example, it's certainly not huge but is getting to be decent-size…

Hi. Nice product!

Let's say I have a repo for an NLP project. One directory contains a few thousand text files. Can I tell Plandex to never ever index and access them? For my use case, I wish projects in this space always asked me before accessing anything. Claude recently decided to install seven Python packages and grabbed full terminal output following installation, which turned out pretty expensive (and useless).

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#75

Post by danenania »

dr_kiszonka wrote: post #74
danenania wrote: post #41

Earlier quoted context omitted.

It's a bit hard to give "typical" costs because it's so dependent on how you use it. The project map size (which scales with overall project size) and the number/size of relevant files are the main drivers of cost, so working in large existing codebases will be a lot more expensive than generating a new app from scratch. Taking Plandex's codebase as an example, it's certainly not huge but is getting to be decent-size…

Hi. Nice product! Let's say I have a repo for an NLP project. One directory contains a few thousand text files. Can I tell Plandex to never ever index and access them? For my use case, I wish projects in this space always asked me before accessing anything. Claude recently decided to install seven Python packages and grabbed full terminal output following installation, which turned out pretty expensive (and useless).

Hi, thanks! Yes, you could either:

- Add that directory to either .gitignore (in a git repo) or a .plandexignore file (which uses gitignore syntax).

- You can switch to a mode where context is not loaded automatically and you choose the files yourself instead (more on this here: https://docs.plandex.ai/core-concepts/autonomy).

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#76

Post by danenania »

killerstorm wrote: post #72
danenania wrote: post #66

Earlier quoted context omitted.

Thanks for trying it! Could you explain in a bit more detail what went wrong for you with sign-in and the pin? Did you get an error message? On OpenRouter vs. OpenAI, see my other comment in this thread ( https://news.ycombinator.com/item?id=43719681 ). I'll try to make this smoother. On Gemini 2.5 Pro: the new paid 2.5 pro preview will be added soon, which will address this. The free OpenRouter 2.5 pro experimental…

When I used `--pin` argument I got an error message along the lines of "not found in the table". I got it working by switching to oss model pack and specifying G2.5P on top. Also works with anthropic pack. But I'm quite disappointed with UX - there's a lot of configuration options but robustness is severely lacking. Oddly, in the default mode out of box it does not want to discuss the plan with me but just jumps to i…

Thanks, I appreciate the feedback.

> Oddly, in the default mode out of box it does not want to discuss the plan with me but just jumps to implementation.

It should be starting you out in "chat mode". Do you mean that you're prompted to begin implementation at the end of the chat response? You can just choose the 'no' option if that's the case and keep chatting.

Once you're in 'tell mode', you can always switch back to chat mode with the '\chat' command if you don't want anything to be implemented.

> And when it's done writing code it aggressively wants me to decide whether to apply -- there's no option to discuss changes, rewind back to planning, etc. Just "APPLY OR REJECT!!!". Even Ctrl-C does not work! Not what I expected from software focused on planning...

This is just a menu to make the commands you're most likely to need after a set of changes is finished. If you press 'enter', you'll return back to the repl prompt where you can discuss the changes (switch back to chat mode with \chat if you only want to discuss, rather than iterate), or use commands (like \rewind) as needed.

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#77

Post by danenania »

throwup238 wrote: post #71
danenania wrote: post #70

Earlier quoted context omitted.

It should already be skipping the email validation step in local mode. Is it showing up for you? I’ll look into the detached mode, thanks!

Yes, it showed up for me, luckily I had the logs open and remembered that was the solution in v1 (it wasn’t documented back then iirc). I git pulled in the same directory I ran v1 in so maybe there’s some sort of left over config or something?

Email pins are disabled based on a LOCAL_MODE environment variable, which is set in the docker-compose config. I'll take a look.

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#78

Post by killerstorm »

danenania wrote: post #76
killerstorm wrote: post #72

Earlier quoted context omitted.

When I used `--pin` argument I got an error message along the lines of "not found in the table". I got it working by switching to oss model pack and specifying G2.5P on top. Also works with anthropic pack. But I'm quite disappointed with UX - there's a lot of configuration options but robustness is severely lacking. Oddly, in the default mode out of box it does not want to discuss the plan with me but just jumps to i…

Thanks, I appreciate the feedback. > Oddly, in the default mode out of box it does not want to discuss the plan with me but just jumps to implementation. It should be starting you out in "chat mode". Do you mean that you're prompted to begin implementation at the end of the chat response? You can just choose the 'no' option if that's the case and keep chatting. Once you're in 'tell mode', you can always switch back t…

Here's what happened:

  1.  It started formulating the plan
 2.  Got error from provider (it seems model set sometime randomly resets to default?!?)
 3.  After I switched to different provider, I want it to continue planning, so I use \continue command
 4.  But when it gets \continue command it starts writing code without asking anything!
 5.  In the end it was still in chat mode. I never switched to tell mode, I just wanted it to keep planning.
Here's an excerpt: https://gist.github.com/killerstorm/ad8afa19b2f55588eb317138...

It went from entry 3 "Made Plan" to entry 4 and so on without any input from my end.

I could not reproduce the second issue this time: I didn't get the same menu and it was more chill.

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#79

Post by bottled_poe »

lightdot wrote: post #40

From the Github page: "curl -sL https://plandex.ai/install.sh | bash" Enticing users to blindly run remote 3rd party code on their machines is IMHO not a proper thing to do. This approach creates a dangerous mindset when it comes to security and good practices in general.

How is this any different to downloading and running a binary?

Re: Show HN: Plandex v2 – open source AI coding agent for large projects and tasks

#80

Post by lsdkfjlkasfj »

danenania wrote: post #8
maxwelljoslyn wrote: post #2

It seems that much like Aider, you use separate models for creating code edits and validating them. That's a win in my book. It seems Claude Code does not do that, which is part of the reason it racks up a (relatively) large bill for long work sessions (that and the cost of sonnet-3.7). I bounce back and forth between Aider, Claude Code, and Simon Willison's LLM tool ("just" a GOOD wrapper for using LLMs at the CLI,…

> It seems that much like Aider, you use separate models for creating code edits and validating them. That's right. To apply edits, Plandex first attempts a deterministic edit based on the edit snippet. In some cases this can be used without validation, and in others a validation step is needed. A "race" is then orchestrated with o3-mini between an aider-style diff edit, a whole file build, and (on the cloud service)…

What are the main differences from aider?
Post reply on HN