#Claude
1 messages · Page 1 of 1 (latest)
Yeah, basically yes.
If you mean using your Claude Pro/Max subscription inside OpenClaw, the normal path is to log in with Claude CLI on that machine and let OpenClaw reuse that auth.
On the ban question: current OpenClaw docs say Anthropic staff told them this usage is allowed again, so OpenClaw treats Claude CLI reuse and claude -p as sanctioned unless Anthropic changes policy later. So it is not treated as some shady workaround right now.
Small nuance: having “extra usage” available still depends on your Anthropic plan/account eligibility, especially for bigger/long-context requests. Connecting Claude CLI is the auth path, but it does not magically unlock more than your account allows.
Docs:
- https://docs.openclaw.ai/providers/anthropic
- https://docs.openclaw.ai/concepts/oauth
- https://docs.openclaw.ai/help/faq
If you want, I can also show the exact safest setup for Claude CLI auth in OpenClaw.
I got a email that said "Hi,
We're offering you a one-time credit for extra usage to your account, equal to your monthly subscription price. Redeem it here by April 17. Once claimed, it’s good for 90 days across Claude Code, Claude Cowork, chat, or third-party harnesses connected to your account.
You’re also now able to pre-purchase bundles of extra usage at up to 30% off. If you ever run past your subscription limits, this is the easiest way to keep going.
One note: as of April 4, third-party harnesses like OpenClaw connected to your Claude account now draw from extra usage instead of from your subscription. If you don't use them, nothing changes. If you do, the credit and bundles above have you covered.
Thanks for building with Claude."
Yep, that email is pretty explicit.
What it means in practice:
- Yes, if you connect OpenClaw through your Claude account auth / Claude CLI path, it should count against extra usage.
- No, that does not look bannable. Anthropic is literally calling out “third-party harnesses like OpenClaw” in the email, so this is being acknowledged as a supported use case.
- Important catch: the one-time credit had to be redeemed by April 17. If you already claimed it, you should be good for 90 days. If not, that promo credit may be gone.
So the short version is:
Claude CLI/account auth = yes, that’s the right connection path for this.
And based on that email, OpenClaw usage through that path is explicitly expected, not sketchy.
If you want, I can give you the exact OpenClaw auth path to make sure it uses Claude account auth and not an Anthropic API key.
So I shouldn't get banned if my auth method is anthropic Claude cli? In openclaw
Right, based on Anthropic’s email and OpenClaw’s current docs, you should not expect a ban for using OpenClaw with official Anthropic Claude CLI auth.
The careful version is:
- OpenClaw supports Claude CLI auth
- Anthropic explicitly mentioned third-party harnesses like OpenClaw
- so this is not being treated as a prohibited workaround
I just would not say “100% impossible forever,” because Anthropic can always change policy later. But as things stand now, this is an allowed/supported path.
If you want the most boring/safest possible setup anyway, use an Anthropic API key. If you want to use your Claude account extra usage, Claude CLI auth is the right path.
I have been talking to openclaw for like 15mins now it has just added the personality layer to its entire file system for memory and my extra usage credits sit at the exact same balance like did not decrees even a cent
That alone does not mean it is not using Claude auth.
A few very normal reasons:
- usage/balance UI can lag
- a lot of what you described, like memory/file updates, can be pretty cheap
- OpenClaw might be using another provider or fallback model for some turns if your config is set that way
If you want to verify instead of guessing, check these two things:
openclaw --version
openclaw models status
What you want to see is that the active/authenticated path is Anthropic via Claude CLI/account auth, not an API key or some other provider.
If you paste the openclaw models status output, I can tell you pretty quickly whether it is actually drawing from your Claude account or not.
No it's using the models for sure I can tell by the way it talks and by the fact that it's the only thing I configured I didn't do anything else