Hacker News Reader

Comment by Flux159 | Hacker News Reader

With how good Claude 3.5 is for coding, I would love if this supported private github repos - just connect my github account and provide a repo link which stays up to date as changes are made instead of having to upload content to a project.

I feel like this will probably be an option down the line - I wonder if it would be possible to replicate this functionality via the API as well, kind of like how there are open source projects attempting to replicate artifacts via the API.

View on HN

You could absolutely get this working via the API right now. Here's how to use my API CLI tool to do that:

    pip install files-to-prompt llm llm-claude-3
    llm keys set claude
    # paste in your Claude API key
    llm models default claude-3.5-sonnet
    files-to-prompt my-github-repo/ | llm --system 'say OK'

And now you can start an interactive chat with Claude 3.5 Sonnet pre-seeded with your files like this:

    llm chat --continue

The --continue will ensure it is a continuation of the previous chat session where you dumped in all those files.

If you want to run against an updated copy of your files, start a new conversation with the files-to-prompt piping again.

+1 - I've been using LLM CLI to do this, and it's great. The best part is that you can swap models without changing your workflow.

I use Claude 3.5 for my daily coding tasks, but used Gemini Pro 1.5 to analyze a new Github repository the other day because it has a much larger context size.

You simply run `llm -m gemini-1.5-pro-preview-0409 "<your question>"`

Any thoughts on how the quality of this compares with something like Greptile or other tools dedicated to trying to do this?

How good is this though? IIRC after a certain amount of lines in uploaded files Claude seemed to "Forget" what I was talking about or get confused.

Is Claude’s API available to individuals. Last time I tried it was only available to businesses

Yea it's open to individuals now: https://www.anthropic.com/api

You may have to add a credit card in order for it to give you an API key, it's been a while since I set mine up.

or use:

    find "$1" -type f \( ! -regex '.*/\..*' \) -print | while IFS= read -r file
    do
      echo "$file"
      echo "---"
      cat "$file"
      echo "---"
    done

I use `find -name '*.py' | xargs tail -n +0 | sed 's/ / /g' | pbcopy` a lot for this! Easy to remember.

This gets expensive really quickly at least for me. The web ui for 20 dollar is dirt cheap in comparison.

With the new Projects feature I was able to hit the message rate limit much faster in Claude, after the API is really the only option until reset time.

I think it’s still a token per time limit. So more files in the project and more artifacts means token usage.

I know people have been doing really cool things writing new projects. But everything I try to do with Sonnet 3.5 with an existing codebase, with its own intricacies like library versions is really bad

I've been trying to generate test coverage with 3.5 on a repo that has existed for years. it can take up to 20 attempts of file -> this is the error i get -> error -> error until i get one that runs successfully. It slightly improves when i give it a file, and also give it other files it successfully generated after 20 attempts and ask it to use the same style and learn from the changes it made previously to make it work

But its very hard to get it done with less than 10 attempts.

Ultimately, I expect them to offer a cloud IDE with integrated AI, like Google's IDX, but with better AI.

IF you are using it for coding you can use cody.dev, which indexes your code and supports Claude 3.5

I second using Cody. It has been great for me. Being able to change models is fantastic.

I bet one problem with supporting repos is that the 200K context can fill up quick. Need to be selective with what files are in the context. If the repo was integrated, and you could select which files are in the current context, that could be good.

I think the first step is to have a .claudeignore file (or maybe we can standardize on .llmignore) to tell it which files you don’t want it to add to the context. You could easily eliminate millions of tokens by ignore node_modules in most JavaScript projects!

You would ignore node modules in any case, but our monorepo has millions of files outside node modules or other generated artifacts, so going per package is still 100k files or so. LLMs don’t like that much yet.

I've helped build https://gptengineer.app/ which works exactly like that. We support Claude 3.5.

Looks exciting! Can’t wait to start using it.

“You are currently nr 27341 on our waitlist”. Oh.

How about a CLI tool that uses Claude's APIs to make changes to the files and do commits? Or, is it a must this is integrated into a web page and/or code editor?

This would also work for local development and then put up a pr, but I feel like that's a temporary thing as well until a model or agent-like workflow is good enough to directly read from a github issue and make the PR directly.

Right now Cursor can already use Claude 3.5 so I feel like that's very close to integrated into a code editor.

This seems to exist in some capacity. It was demo'd in the following video: https://x.com/alexalbert__/status/1803804679891255688

I haven't been able to determine if it is using an existing agent like Aider, or something else.

Only problem is when the generated code exceeds like 300 lines or something, it stops and says "it hit maximum length allowed"

Have you tried it for writing documentation ?

Reply on news.ycombinator.com