It looks like the key idea here is that a GitHub repo can have a gemini-extension.json file which specifies some MCP servers and a GEMINI.md file, then you can tell Gemini CLI to add that repo and it will fetch and configure those details.
"The extensions listed here are sourced from public repositories and created by third-party developers. Google does not vet, endorse, or guarantee the functionality or security of these extensions. Please carefully inspect any extension and its source code before installing to understand the permissions it requires and the actions it may perform. "
Extensions are always a trust challenge, but the high value of AI systems means that I expect we'll see a very high volume of attacks in the near future.
Cool, but at the same time, it feels overwhelming: so many different CLI or IDE tools, so many extension points. It will be fascinating to see how this all shakes out.
The last time I was locked out of Claude Code I gave gemini cli a try. 10 minutes and lots of weird scrolling and screen flashing later, and it kept trying to add and remove the same function from the same file over and over, in a loop, until it exhausted its credits and demanded I upgrade.
What it excels at (because of its massive 1M token context) is code reviews. As a reference the French version of The Three Musketeers is about 700k tokens. A ~440 odd page book.
You can shove any codebase at it and ask questions about how it works and it's all in its context at the same time with no RAG magic.
My recent experience of Gemini CLI in comparison to Claude code is very negative. For a complex task of trying to migrate some code base from CPU-only to GPU computations Claude was really helpful, and allowed me to conduct a range of experiments and converge towards sensible architecture. In the same time I simply was not able to get anything useful out of Gemini. Some code was written, but it was not tested, was failing or timing out and after repeated prodding Gemini got into loops or internal errors.
Gemini is really bad with tool calling, doesn't matter if you use it with Gemini CLI or with OpenCode, its just that the model itself is not a agentic model. It can be okay if only use it for asking stuff about the codebase, but not for doing agentic stuff. It'll waste large parts of its context window just failing to do tool calling properly.
If there's value in a human using Figma, there's value in being able to give a "bot" instructions to do whatever a person might. e.g. "Gemini, update all the Figma layouts with this new text."
I’ve regularly used prompts along the lines ‘please modify the UI tor x feature to look like the selected Figma frame’. It’s ideal for a designing developer who wants to be much more specific about the user experience.
Gemini is about 10x cheaper per token. But for some reason it's using 8 times more input tokens than CC. They also have this thing called cached tokens, which is much cheaper than not cached tokens. It's a hot cache of your context on Google side, cached automatically.
So at the end of the day you don't know how much you'll pay.
Models
Google is good for very complex topics and when the conversation is short. But both models are great. I prefer Claude and Sonnet 4.5 is great all around
CLI tools
Gemini cli is at it's very early days. Doesn't support hooks or subagents. Often runs into loops it can't break out from, essentially gets stuck but you still pay for the tokens.
Claude is just great. Allows you to write complex workflows they way they are supposed to be written. Handles hooks and subagents. MD file can reference another MD file, so you can DRY your files.
Nested plan mode works weird, sometimes the agent gets stuck if it asks for plan approval and thinks it's executing it, but displays nothing... So plan mode is not fully supported in subagents.
A nice thing is that .Claude directory is automatically understood by codex or cursor, you should be able to run your Claude command using openai models via codex or maybe even other providers via Cursor.
Summary
Overall Claude is the best all around, but the tokens are crazy expensive and the subscription model is a joke. You don't know how many tokens you can use when you're subscribed, but it's 'something', and last week they changed the limits, it's suddenly half of 'something'...
Using both. Sometimes Claude makes stupid mistakes, such as skipping the provided document. Overall, Claude gives better enterprise-grade suggestions and code, which Gemini sometimes misses. However, if you have in-depth documentation of the scope and implementation approach, Gemini performs better.
My primary tool is Gemini Code Assist. Claude is used to create the draft implementation approach and for a final sanity check of the code, as well as to propose enhancements for production readiness. This combination has worked well for me. Since Claude is expensive and Gemini is more affordable, it also makes economical sense.
I usually provide a well defined scope and detailed implementation approach, with the whole project split into submodules and the scope and implementation approach is again refined for each modules. In my experience, the programming language also matters, results are often better when using statically typed languages. We use C# and front end is always developed without AI.
The only compelling reason for me to try Gemini CLI opposed to Claude Code is that Gemini CLI supposedly supports ollama, which could potentially open up avenues for locally hosted agentic coding (lower quality, but private).
When I attempted to connect to my local ollama instance (~3 months ago) the Gemini CLI config and instructions just didn't work and weren't intuitive, so I gave up.
26 comments
[ 2.4 ms ] story [ 48.0 ms ] threadHere's a commit I found adding one of these for the Google Maps platform: https://github.com/googlemaps/platform-ai/commit/95c4efdb43c...
Extensions are always a trust challenge, but the high value of AI systems means that I expect we'll see a very high volume of attacks in the near future.
What it excels at (because of its massive 1M token context) is code reviews. As a reference the French version of The Three Musketeers is about 700k tokens. A ~440 odd page book.
You can shove any codebase at it and ask questions about how it works and it's all in its context at the same time with no RAG magic.
I used to use Gemini occasionally with cursor 6 months ago and it wasn’t this bad. So I’m not sure what exactly caused it to be that bad.
Why would I want to do that?
I feel no desire to switch or learn a new thing but I'm wondering if people feel like it's on par with CC or Codex or behind.
Gemini is about 10x cheaper per token. But for some reason it's using 8 times more input tokens than CC. They also have this thing called cached tokens, which is much cheaper than not cached tokens. It's a hot cache of your context on Google side, cached automatically. So at the end of the day you don't know how much you'll pay.
Models
Google is good for very complex topics and when the conversation is short. But both models are great. I prefer Claude and Sonnet 4.5 is great all around
CLI tools
Gemini cli is at it's very early days. Doesn't support hooks or subagents. Often runs into loops it can't break out from, essentially gets stuck but you still pay for the tokens.
Claude is just great. Allows you to write complex workflows they way they are supposed to be written. Handles hooks and subagents. MD file can reference another MD file, so you can DRY your files.
Nested plan mode works weird, sometimes the agent gets stuck if it asks for plan approval and thinks it's executing it, but displays nothing... So plan mode is not fully supported in subagents.
A nice thing is that .Claude directory is automatically understood by codex or cursor, you should be able to run your Claude command using openai models via codex or maybe even other providers via Cursor.
Summary
Overall Claude is the best all around, but the tokens are crazy expensive and the subscription model is a joke. You don't know how many tokens you can use when you're subscribed, but it's 'something', and last week they changed the limits, it's suddenly half of 'something'...
My primary tool is Gemini Code Assist. Claude is used to create the draft implementation approach and for a final sanity check of the code, as well as to propose enhancements for production readiness. This combination has worked well for me. Since Claude is expensive and Gemini is more affordable, it also makes economical sense.
I usually provide a well defined scope and detailed implementation approach, with the whole project split into submodules and the scope and implementation approach is again refined for each modules. In my experience, the programming language also matters, results are often better when using statically typed languages. We use C# and front end is always developed without AI.
I use Gemini Standard Tier and Claude Pro Tier.
When I attempted to connect to my local ollama instance (~3 months ago) the Gemini CLI config and instructions just didn't work and weren't intuitive, so I gave up.
But nice of them to try wrestling a file into your repo named “gemini something something”.. can’t knock them for having a go.
This one seems to showcase a bunch of the "extension" features, including a custom MCP for dealing with file line numbers.
Feedback, bug reports, and ideas are all welcome on the GitHub repo's issues tab. Happy to answer any questions here too.
https://github.com/google-gemini/gemini-cli/issues/4310
Well I guess not...