Is there any other way to do it though? Even if they implemented some form of auth logic, since it's all client side anyone could spoof it. The only real way to distinguish Claude Code from non-Claude Code is behavioral analysis (e.g. system prompt, set of tools, etc.). Or engage in a cat and mouse game of increasingly obfuscated challenge/auth.
Or call the tool "Read" and it works, according to an issue comment.
But actually the solution is checking out how the official client does it and then doing the same steps, though if people start doing this then Anthropic will probably start making it more difficult to monitor and reverse engineer.
It might not matter, as some people have a lot of expertice in this, but people might still get the message and move away to alternatives.
Then they'd start pinning certs and hiding keys inside the obfuscated binary to make traffic inspection harder?
And if an open source tool would start to use those keys, their CI could just detect this automatically and change the keys and the obfuscation method. Probably quite doable with LLMs..
Aren't Anthropic in control of all the legitimate clients? They can download a new version, possibly automatically.
I believe the key issue here is that the product they're selling is all-you-can-eat API buffet for $200/month. The way they manage this is that they also sell the client for this, so they can more easily predict how much this is actually going to consume tokens (i.e. they can just put their new version of Claude Code to CI with some example scenarios and see it doesn't blow out their computing quota). If some third party client is also using the same subscription, it makes it much more difficult to make the deal affordable for them.
As I understand it, using the per-token API works just fine and I assume the reason people don't want to use it because it ends up costing more.
Here’s the thing though…if you read enough of it, you’re gonna start using it a lot more often. It’s not just AI slop, it’s fundamentally rewiring how we as a society think in real time! It’s the classic copycat AI mannerisms cried wolf
Problem!
Stencil has a way better DX. Especially if you’re doing a whole library of components. You pay a slightly higher perf tax for their auto loader and light vdom. Lit is the way to go if you have opinions and want no magic
"GO: A Complete Introduction to the Game" by Cho Chikun is a good place to start. Then just start playing games. You can be auto-matched on online-go.com with players of close to your strength (and you can request a handicap in the settings as well. You are what is known as 30 kyu player (think "white belt" in karate). The great thing about go is that the handicap system allows you to play people of greatly varying skill and have a chance at beating them. Right now you could probably play up to a 21 kyu player with a 9 handicap (19x19 board) and still be on equal footing [1].
I can't stress enough the need to just play games. Have fun. If you don't enjoy it you won't keep doing it, so don't stress out about "doing the right thing." Just try stuff and see if it works. If it doesn't, oh well. Don't do that next time. I know I said "don't just try things" above, but that was specifically in learning to read life-and-death problems. At your level you should just be having fun playing games right now. Life-and-death problems can be boring (but necessary) for a lot of players so the important thing right now is that you enjoy playing. The other stuff will come along naturally as you develop the desire to get better.
Oh, and don't play against AI (even weak ones). They don't play like normal humans and will just confuse you more than help. At much higher levels, AI has been very helpful in finding new tactics and strategies that more optimal, but at your level, playing other humans will get you better much faster.
I’ve become obsessed with pattern languages lately. If you read Timeless Way of Building, you realize we got it all wrong. What everything thinks of as a “pattern” is missing the point entirely. It’s the pattern language that has power.
He even calls out atomic design and all of the flaws with it—no two atoms are actually identical. So the result is we have these rigid design systems that discourage variation because, imo, we can’t be bothered to define the things that make something alive.
Well let me tell you
https://github.com/anomalyco/opencode/blob/dev/packages/open...
You literally send your first message “you are Claude code”
The fact that this ever worked was insane.
Headline is more like anthropic vibes a bug and finally catches it.