artdigital

Alright, I wanted to give Jules another fair try to see if it improved, but it's still terrible.

- It proposed a plan. I gave it some feedback on the plan, then clicked approve. I came back a few minutes later to "jules is waiting for input from you", but there was nothing to approve or click, it just asked "let me know if you're happy with the plan and I'll get started". I told Jules "I already approved the plan, get started" and it finally started

- I have `bun` installed through the environment config. The "validate config" button successfully printed the bun version, so it's installed. When Jules tries to use bun, I get `-bash: bun: command not found` and it wastes a ton of time trying to install bun. Then bun was available until it asked me for feedback. When I replied, bun went missing again. Now for whatever reason it prefixes every command with "chmod +x install_bun.sh && ./install_bun.sh", so each step it does is installing bun again

- It did what I asked, then saw that the tests break (none were breaking beforehand, our main branch is stable), and instead of fixing them it told me "they're unrelated to our changes". I told it to fix everything, it was unable to. I'm using the exact same setup instructions as with Copilot Agent, Codex and Claude Code. Only Jules is failing

- I thought I'll take over and see what it did, but because it didn't "finish", it didn't publish a branch. I asked it to push a branch, it started doing something and is now in "Thinking" for a while. Seems to be running tests and lint again which are failing. But eventually it published the branch.

At this point I gave up. I don't have time to debug why bun is missing when in the env configuration it is available, or why it vanished in between steps, or figure out why only jules isn't able to properly run our testsuite. It took forever for a relatively small change, and each feedback iteration is taking forever.

I'm sure it'll be great one day, and I'll continue to re-visit it, but for now I'll stick with the other 3 when I need an async agent

And Gemini CLI github action (this project) runs again in a VM (github action runner) on a separate checkout of the code. This is what OP meant with multiple coding agents.

And this can't authenticate the same way the normal gemini cli does, it needs an API key from the looks of it, so free, standard and enterprise plans through OAuth currently don't work for authentication, just the free tier of the Google AI Studio, which is different than gemini-cli free tier, and has way tighter rate limits.

> Curious to try this against the Github (website) Agent. The website Agent is definitely dumber than the vscode agent (because it has to spend 20 minutes figuring out how to build and start my monorepo apps) but on the flip side, it doesn't take up my computer and thus any value it creates is additive.

Yeah that's on you. Add a `copilot-instructions.md` file and configure the `copilot-setup-steps.yml` workflow to setup your environment. Both are supported more or less since Copilot Agent got released (though in "preview")

Most agents read `AGENTS.md`, I just symlink it to CLAUDE.md, and do the same for GEMINI.md

I wonder why they call this `gemini cli`, it's not really a CLI anymore when it's primarily used through GitHub, is it?

Why not follow Claude Code naming with this and just call it `gemini github action` or `run gemini`?

Yeah agree. I currently don't subscribe to any of the Gemini things because I'm too much in the Claude ecosystem and can't justify another sub. But if I were to subscribe to something, it would be the Google Developer Premium plan. Mostly for Gemini CLI, but that alone is currently still way too rate-limited and not worth it.

Maybe if they add Jules and some version of Gemini to it (without the Gdrive storage and YT premium stuff), I would get onto it.

Having Jules in the normal consumer Gemini subscription makes no sense. It's very clearly a dev product, so it should be bundled with other dev products, so through GCP with Code Assist or Google Developer Premium. It feels completely out of places in the list of features on https://one.google.com/ai

You're forgetting the 'Google Developer Premium' subscription ($300/y) which also bundles Gemini Code Assist / CLI, some Vertex credits, but none of the other Gemini things

One can make an argument that other Gemini stuff shouldn't be in there because it's not dev related, but Jules at least should

I tried Jules multiple times during the preview. Almost every week once, and it’s pretty terrible. Out of all the cloud coding assistants, it’s the worst. I honestly thought it was just an experiment that got abandoned and never expected it to actually become a real product, similar to how GH Copilot Spaces was an experiment and turned into Copilot agent.

It does what it wants, often just “finishes” a task preemptively and asking follow ups does nothing besides it rambling for a bit, the env sometimes doesn’t persist and stuff just stops working. For a while it just failed completely instantly because the model was dead or something

Out of the dozen times I tried it, I think I merged maybe one of its PRs. The rest I trashed and reassigned to a different agent.

My ranking

- Claude Code (through gh action), no surprise there

- ChatGPT Codex

- GitHub Copilot Agent

- Jules

I will try it again today to see if the full release changed anything (they give 3 months free trial for previous testers) but if it’s the same, I wouldn’t use or pay for Jules. Just use Codex or GitHub agent. Sorry for the harsh words

Q CLI is great. It’s basically Claude models but pretty much unlimited and only for $20

Not as polished as Claude Code but also a bit price difference

I never have tool calls failing on macOS. What tool calls are failing for you?

Wanted to give it a proper try but because it's so overloaded, it almost instantly downgraded me to 2.5-flash. The vscode extension didn't even load, it just gave me a rate limit warning directly after logging in

Tried upgrading to the Standard plan through Google Cloud with the hope that it would allow me to do more, but after switching my account to the now-paid account, it still almost instantly downgraded me to 2.5-flash

For the times when I was able to use 2.5-pro, the output has been very good. But the moment it switches to flash, the quality goes down by about 80% and I would never let it code on anything

How can I subscribe to this? From the comments here it sounds like it's NOT bundled into the Gemini Pro subscription, but into "Gemini Code Assist Standard".

I can't find any way to upgrade to a paid plan, is this even possible for individuals, or is it just "free or enterprise"?

/Edit: Okay I went through the Gemini docs. I found that in Google Cloud you can enable Gemini Code Assist Standard and Enterprise for the account

- Standard is $19.00/mo

- Enterprise is $45.00/mo

Difference between the 2 editions: https://cloud.google.com/products/gemini/pricing

/Edit2: Found the actual management screen: https://codeassist.google.com/overview

Not a desktop app, but Gemini has “Jules” which is their autonomous coding agent

You hook up your GitHub repo and it’ll clone it, setup the env and then work on the task you give it

Doesn’t it require you to create a new chat?

Add a `product.json` and set the extension gallery URL through that

``` { "extensionsGallery": { "serviceUrl": "https://marketplace.visualstudio.com/_apis/public/gallery", "itemUrl": "https://marketplace.visualstudio.com/items", "cacheUrl": "https://vscode.blob.core.windows.net/gallery/index", } } ```

https://github.com/getcursor/cursor/issues/2195

This is technically against the ToS of the extension gallery though

Windsurf users can still plug their own Anthropic key and continue using the models. It’s Windsurf subscribers (eg OpenAI customers) that use the models through the Windsurf service (through their servers as proxy, that’s now OpenAI) are getting cut off

I don’t see how this is irrelevant. Windsurf is a first party product of their most direct competitor. Imagine a car company integrating the cloud tech of a different manufacturer

> This feels like a cheap trick to drive users towards Claude Code

How did you come to this conclusion? It’s very much like he remarked: OpenAI acquired Windsurf, OpenAI is Anthropics direct competitor.

It doesn’t make strategic sense to sell Claude to OpenAI. OpenAI could train against Claude weights, or OpenAI can cut out Anthropic at any moment to push their own models.

The partnership isn’t long lasting so it doesn’t make sense to continue it.

Swift has all of that already answered. Non Xcode swift with SwiftPM is solid

I’m not a fan of the 20 tool limit unless you use the Max option which costs you 1 credit for each and every tool call + message. Seems like an artificial limit and it always rips me out

Interesting. For me Cline and Roo are king. I would use them exclusively if I could afford it. With Copilot Pro+ it goes a long way but still ends in rate limits down the road

You can change it back to use the normal Microsoft marketplace

I’m now also on Max but had to upgrade to the 200 plan in about a week of daily rate limits

The $200 plan so far has been fine. Only had it once that it looked like I might hit the limit soon, but that was a very heavy refactoring task

I’m on Pro+ and get rate limited heavily. 1-2 hours of semi heavy use and the brakes kick in. I can’t stay productive in it because this always rips me out

It’s getting better fast. It’s still needs refining but I have no issues spending the entire day in it

Hearing a company like Apple is using swift on the server side changed my view on server Swift

I’ll definitely take a look again, great to see it becoming mature enough to be another viable option

Oh jeez yes. I completely forgot that this was a thing. It’s tendency to do completely different things “while at it” was ridiculous

There are writeups about this. The sites score each move, it is extremely unlikely to pick superhuman moves multiple times in a row. Once or twice maybe, but not most of the times.

iPhones yes, Android no. Most Android devices sold outside of Japan can’t use FeliCa so you have to get the Japan model for osaifu-keitai stuff

I find Windsurf almost unusable. It’s hard to explain but the same models in Zed, Windsurf, Copilot and Cursor produce drastically worse code in Windsurf for whatever reason. The agent also tends to do way more stupid things at random, like wanting to call MCP tools, creating new files, then suddenly forgetting how to use tools and apologizing a dozen times

I can’t really explain or prove it, but it was noticeable enough to me that I canceled my subscription and left Windsurf

Maybe a prompting or setting issue? Too high temperature?

Nowadays Copilot got good enough for me that it became my daily driver. I also like that I can use my Copilot subscription in different places like Zed, Aider, Xcode

Gemini 2.5 pro is great, but also VERY expensive with non opaque cost insights

Just recently a lot of people (me included) got hit with a surprise bill, with some racking up $500 in cost for normal use

I certainly got burnt and removed my API key from my tools to not accidentally use it again

Example: https://x.com/pashmerepat/status/1918084120514900395?s=46