There seem to be a ton of people who don't understand how subscription services work. Every single one of them oversells their capacity. The power users that use the services a lot are subsidized by those who don't use it as much, which tends to be the vast majority of the user base. OpenClaw is an autonomous power user. The growing adoption of this walking attack surface was either going to A) cause the cost of Claude to go up or B) get banned to protect the price of the service for actual users.
I suspect people are misdiagnosing the root cause of why Anthropic is doing this a bit.
I don't think this is particularly about the financial impact of people using OpenClaw - they can adjust the amount of tokens in a subscription quite easily.
I think the root cause is that Anthropic is capacity constrained so is having to make choices about the customers they want to serve and have chosen people who use Claude Code above other segments.
We know Anthropic weren't as aggressive as OpenAI through 2025 in signing huge capacity deals with the hyperscalers and instead signed smaller deals with more neo-clouds, and we know some of the neo-clouds have had trouble delivering capacity as quickly as they promised.
We also know Claude Code usage is growing very fast - almost certainly faster since December 2025 than Anthropic predicted 12 months ago when they were doing 12-month capacity planning.
We know Anthropic has suffered from brown-outs in Claude availability.
Put this all together and a reasonable hypothesis is that Anthropic is choosing which customers to service rather than raising prices.
My answer to this is simply rolling back to the pro plan for interactive usage in the coming month, and forcefully cutting myself over to one of the alternative Chinese models to just get over the hump and normalise API pricing at a sensible rate with sensible semantics.
Dealing with Claude going into stupid mode 15 times a day, constant HTTP errors, etc. just isn't really worth it for all it does. I can't see myself justifying $200/mo. on any replacement tool either, the output just doesn't warrant it.
I think we all jumped on the AI mothership with our eyes closed and it's time to dial some nuance back into things. Most of the time I'm just using Opus as a bulk code autocomplete that really doesn't take much smarts comparatively speaking. But when I do lean on it for actual fiddly bug fixing or ideation, I'm regularly left disappointed and working by hand anyway. I'd prefer to set my expectations (and willingness to pay) a little lower just to get a consistent slightly dumb agent rather than an overpriced one that continually lets me down. I don't think that's a problem fixed by trying to swap in another heavily marketed cure-all like Gemini or Codex, it's solved by adjusting expectations.
In terms of pricing, $200 buys an absolute ton of GLM or Minimax, so much that I'd doubt my own usage is going to get anywhere close to $200 going by ccusage output. Minimax generating a single output stream at its max throughput 24/7 only comes to about $90/mo.
This is slightly different from what OpenCode was banned from doing; they were a separate harness grabbing a user’s Claude Code session and pretending to be Claude Code.
OpenClaw was still using Claude Code as the harness (via claude -p)[0]. I understand why Anthropic is doing this (and they’ve made it clear that building products around claude -p is disallowed) but I fear Conductor will be next.
Skimming through the comments, it feels like I am reading the same message over and over. I agree with some comments that are pointing out the issue with Anthropics capacity constraints and when Subscription vs Api is appropriate.
I would like to point out something else. I have Z.ai subscription and they have a dashboard on my usage.
When trying out Openclaw a while ago, I noted something worrying. Its constantly consuming tokens, every single hour during the day, it consumed tokens. I could see over a period of 30 days, token usage would climb and climb and climb and then shrink to bottom again, as if Openclaw did a context window compaction.
Note, this usage was happening even though I wasn’t using it. It were always running and doing something in the background.
I believe its their Heartbeat.md mechanism. By default it’s set to run every half an hour. I changed it to twice a day, that was enough to me.
I can imagine if thousands of users where connecting their Openclaw instance with default config to Claude with the latest and greatest Opus model, that must’ve felt a bit.
People in the comments are, in my opinion, overcomplicating this and making it more philosophical than it needs to be. The reason for their decision is dead simple: there aren’t enough GPUs, so they have to cut access somewhere, and they’re starting with claw.
It’s really that straightforward. If tomorrow they decide GPUs are better allocated to enterprise use, they could start removing the $20 plan just as quickly overnight, the same way they did tonight.
There are going to be a lot of tools coming soon that are "agent-agnostic", i.e. can run on CLIs including Claude Code. I am personally experimenting with using a combo of MCP + custom UI layer to provide custom tools with bespoke UX and thus turn Claude Code (or any other CLI agent for that matter) into whatever I want. I wonder how they'll deal with that.
For a good existing example developed by a known company, check Cline Kanban: https://cline.bot/kanban
They don't have the MCP-bundling idea that I'm experimenting with, however.
I think the usage patterns of a lot of harnesses are pushing against their planned capacity. I would say they can certainly explain themselves a lot better.
Last week Z.ai coding plan was unusable due to a lot of people abusing the coding plan with OpenClaw. This can be verified: https://openrouter.ai/z-ai/glm-5-turbo
OpenClaw managed to burn 2.46 trillion tokens just in the last 30 days.
I'm not even gonna judge why someone needs an AI Assistant running 24/7, the core issue is that coding plans are being ruined because they're not paying for ridiculous amount of tokens burned.
Anthropic is actually making the right decision: You want a LOT of tokens for your 24/7 agent? Ok, just use the API and pay for your tokens.
I enjoy paying for a sub that I actually use to code, and what we pay today is not even enough to cover the costs of running AI servers.
Big Giant Million Dollar Question: Where does having Openclaw using Claude Code via ACP fall? It's using the Claude Code harness, not the model directly.
If you are not aware, ACP creates a persistent session for steering rather than using the models directly.
Ok. Someone explain to me why they would f themselves this hard with software engineers when they are absolutely winning. This just seems like a bad move.
Is it infrastructure? Are they unable to control costs?
Everyone else is spending like money is water to try to get adoption. Claude has it and is dialing back utility so that its most passionate users will probably leave.
Claude is a UNIX command line tool with an SDK. Yes there's an interactive mode, but it can be invoked as a normal utility too, and piped to other tools and so on.
In that context, I don't understand the difference between a "third party harness" and a shell script.
I don't understand exactly what is being banned. I have a vibe coded context manager + chat thread UI that I use to manage multiple claude code cli sessions simultaneous. Is this allowed? If not how would this get identified vs other cli usage? How is this different than openclaw?
Can they actually realistically do this? Nothing technical can stop a client from masquerading as another, and with the right level of dedication, this wouldn't be very hard to do. And since they're mostly targeting power users, seems like they're barking up the wrong tree. Have I missed something?
Yah well I'll be downgrading my subscription to the $20/month plan for the light chats I have with AI outside of using custom harnesses and will figure out a better provider for the agentic tooling.
I have 0 problem with this. Everybody who was using a Pro subscription with OpenClaw should of expected to be living on borrowed time. The more Anthropic can do to keep the Pro subscriptions at their current price point, the better. It is the best deal in tech imho.
That's why I am using Codex. I slightly prefer Claude in terms of code quality, but it's close, but not being able to use my subscription with other CLIs and apps ruins Claude for me.
1. Make a better product/alternative to Openclaw and start eating their userbase. They hold the advantage because the ones "using their servers too much" are already their clients so they could reach out and keep trying to convert. Openclaw literally brought them customers at the door.
2. Do everyone royally and get them off their platform - with a strong feeling of dislike or hatred towards Anthropic.
Let's see how 2 goes for them. This is not the space to be treating your clients this way.
I am actively using ohmypi harness which is based on pi-mono which I believe is within OpenClaw, I don't personally use OpenClaw but I suspect that I will be affected. The reason that I use ohmypi is because I can extend it and put guardrails specific for our company and myself (those are different from SKILLs and more sophisticated than the hooks) + I like the ability to start "tasks" with faster models like gpt5.4-mini for certain tasks and overall have the multi-model capabilities, now all of this seems impossible. I have the $20 sub from OpenAI and it seems that the usage is similar to the $100 plan by Anthropic, I am extensively using GPT5.4 to review and sometimes code along with Opus, right now it seems to me that OpenAI is winning, I can just go with the $200 unlimited usage by OpenAI and use 5.4/5.4mini for everything. On top of that the Chinese models are really capable at the moment, I've tried StepFun and it's really good. Seems to me that Anthropic is sabotaging themselves with those moves. But it is what it is, the cycle of model switching has begun again, I strongly believe that in 2-3 months they will revert that and we will switch models again. :D
> To make the transition easier, we’re offering a one-time credit for extra usage equal to your monthly subscription price. Redeem your credit by April 17. We’re also introducing discounts when you pre-purchase bundles of extra usage (up to 30%).
The Anthropic casino wants you to continue gambling tokens at their casino only on their machines (Claude Code) only by giving more promotional offers such as free spins, $20 bets and more free tokens at the roulette wheels and slot machines.
But you cannot repurpose your subscription on other slot machines that are not owned by Anthropic and if you want it badly, they charge you more for those credits.
The same reason I would not use a proprietary text editor applies to harnesses. It's enough of a constraint to use a proprietary service, for me the line is at the tooling. Sunk cost and all it's things.
Reality is Ant can supply X tokens and they see demand for 10*X tokens. So they’ll charge whatever the top 10% of users are willing to pay, and slowly degrade the value of the subscriptions until everyone has moved to another supplier or migrated to the 10% price point. The draconian ToS that they sometimes enforce is their mechanism to degrade subscription value over time. Expect agent-sdk to be next on the chopping block, moving from oauth supported to api only. When they switch it they will rightly point out the docs never explicitly said it was allowed.
I had the same problem. Built a Telegram gateway that routes through claude -p (the CLI, still covered by subscription). Self-managing triggers, voice, single Python file. MIT licensed: https://github.com/Kenny1338/claude-telegram-gateway
This has been coming for weeks ever since Anthropic changed their terms and conditions “just a tidy up” - when that happened I took everything I had done in open claw and migrated into Claude code with /loop and tbh I’m happier because I can see in terminals what is happening rather than the odd slack message here or there and I can also use slack to receive messages.
Thanks openclaw for getting me ahead, I’ve taken that and am in Claude code again.
You couldn't make me a happier claw^Hm over this. I am running 3-6 simultaneous agents at once and I have trouble breaking 50% weekly usage with a max plan. What these people are doing is just sloppy engineering. OTOH if you use Claude Code to make code changes, then run that code, the max plan remains more or less free beer for as long as it remains free beer for all the reasons cited elsewhere in this thread.
I am genuinely curious about OpenClaw's continuing allure. I understood it way back then, when Claude Cowork did not have channels and scheduled tasks. But now? Has Claude not become a sane replacement for OpenClaw? I can see that it's fun to play with OpenClaw and non-SOTA providers, but why would anyone run OpenClaw on a Claude Code subscription?
I didn't even realize you could connect a standard subscription to OpenClaw in the first place. It seems like you would run into limits rather quickly, which would degrade the experience quite badly.
Anthropic's current business model is to sell access to their tools to subscribers at a loss. Users maxing out their $200/month plan can realistically cost Anthropic $500-600 in actual compute costs.
Anthropic is okay with this right now because they want to amass as many users as they can, and eventually hope that GPUs will increase in power and efficiency, and their LLMs will become more efficient as well. They can eventually profit off of their current pricing, or with modest price increases, if that comes to fruition.
But letting OpenClaw wake up every 30 minutes and start sending requests is a surefire way to max out your weekly limits, and that certainly isn't something Anthropic planned for.
This is a useful forcing function for distinguishing two architectural categories of Claude Code adjacent tooling. Tools that route your API traffic through a third-party harness are definitionally dependent on Anthropic's policy toward that harness. Tools that run locally and integrate via MCP, without touching the API subscription path at all, are outside this restriction entirely because they are just another tool in your local environment. The local-first architecture was always the right one for teams with compliance or privacy requirements. This week is a good illustration of why it is also the right one for teams that just want to avoid dependency on a vendor-intermediary relationship they cannot control.
Is anyone even getting anything out of a $20/mo sub for Anthropic?
I'm doing a side-by-side with GPT-5.4 for $20/mo and Sonnet for $20/mo and I can tell you that all my 5 hour tokens are eaten in 30 minutes with Claude. I still haven't used my tokens for OpenAI.
Code quality seems fine on both. Building an app in Go
827 comments
I don't think this is particularly about the financial impact of people using OpenClaw - they can adjust the amount of tokens in a subscription quite easily.
I think the root cause is that Anthropic is capacity constrained so is having to make choices about the customers they want to serve and have chosen people who use Claude Code above other segments.
We know Anthropic weren't as aggressive as OpenAI through 2025 in signing huge capacity deals with the hyperscalers and instead signed smaller deals with more neo-clouds, and we know some of the neo-clouds have had trouble delivering capacity as quickly as they promised.
We also know Claude Code usage is growing very fast - almost certainly faster since December 2025 than Anthropic predicted 12 months ago when they were doing 12-month capacity planning.
We know Anthropic has suffered from brown-outs in Claude availability.
Put this all together and a reasonable hypothesis is that Anthropic is choosing which customers to service rather than raising prices.
Dealing with Claude going into stupid mode 15 times a day, constant HTTP errors, etc. just isn't really worth it for all it does. I can't see myself justifying $200/mo. on any replacement tool either, the output just doesn't warrant it.
I think we all jumped on the AI mothership with our eyes closed and it's time to dial some nuance back into things. Most of the time I'm just using Opus as a bulk code autocomplete that really doesn't take much smarts comparatively speaking. But when I do lean on it for actual fiddly bug fixing or ideation, I'm regularly left disappointed and working by hand anyway. I'd prefer to set my expectations (and willingness to pay) a little lower just to get a consistent slightly dumb agent rather than an overpriced one that continually lets me down. I don't think that's a problem fixed by trying to swap in another heavily marketed cure-all like Gemini or Codex, it's solved by adjusting expectations.
In terms of pricing, $200 buys an absolute ton of GLM or Minimax, so much that I'd doubt my own usage is going to get anywhere close to $200 going by ccusage output. Minimax generating a single output stream at its max throughput 24/7 only comes to about $90/mo.
OpenClaw was still using Claude Code as the harness (via claude -p)[0]. I understand why Anthropic is doing this (and they’ve made it clear that building products around claude -p is disallowed) but I fear Conductor will be next.
[0]: See “Option B: Claude CLI as the message provider” here https://docs.openclaw.ai/providers/anthropic#option-b-claude...
I would like to point out something else. I have Z.ai subscription and they have a dashboard on my usage.
When trying out Openclaw a while ago, I noted something worrying. Its constantly consuming tokens, every single hour during the day, it consumed tokens. I could see over a period of 30 days, token usage would climb and climb and climb and then shrink to bottom again, as if Openclaw did a context window compaction.
Note, this usage was happening even though I wasn’t using it. It were always running and doing something in the background.
I believe its their Heartbeat.md mechanism. By default it’s set to run every half an hour. I changed it to twice a day, that was enough to me.
I can imagine if thousands of users where connecting their Openclaw instance with default config to Claude with the latest and greatest Opus model, that must’ve felt a bit.
It’s really that straightforward. If tomorrow they decide GPUs are better allocated to enterprise use, they could start removing the $20 plan just as quickly overnight, the same way they did tonight.
For a good existing example developed by a known company, check Cline Kanban: https://cline.bot/kanban
They don't have the MCP-bundling idea that I'm experimenting with, however.
https://news.ycombinator.com/item?id=46936105 Billing can be bypassed using a combo of subagents with an agent definition
> "Even without hacks, Copilot is still a cheap way to use Claude models"
20260116 https://github.blog/changelog/2026-01-16-github-copilot-now-...
https://github.com/features/copilot/plans $40/month for 1500 requests; $0.04/request after that
https://docs.github.com/en/copilot/concepts/billing/copilot-... Opus uses 3x requests
I think the usage patterns of a lot of harnesses are pushing against their planned capacity. I would say they can certainly explain themselves a lot better.
OpenClaw managed to burn 2.46 trillion tokens just in the last 30 days.
I'm not even gonna judge why someone needs an AI Assistant running 24/7, the core issue is that coding plans are being ruined because they're not paying for ridiculous amount of tokens burned.
Anthropic is actually making the right decision: You want a LOT of tokens for your 24/7 agent? Ok, just use the API and pay for your tokens.
I enjoy paying for a sub that I actually use to code, and what we pay today is not even enough to cover the costs of running AI servers.
AKA when you fully use the capacity you paid for, that's too much!
If you are not aware, ACP creates a persistent session for steering rather than using the models directly.
Is it infrastructure? Are they unable to control costs?
Everyone else is spending like money is water to try to get adoption. Claude has it and is dialing back utility so that its most passionate users will probably leave.
I don’t understand this move.
In that context, I don't understand the difference between a "third party harness" and a shell script.
How are they even detecting OpenClaw?
I switched OpenClaw to MiniMax 2.7. This combined with Claude over telegram does enough for me.
OpenClaw used to burn through all my Claude usage anyway.
1. Make a better product/alternative to Openclaw and start eating their userbase. They hold the advantage because the ones "using their servers too much" are already their clients so they could reach out and keep trying to convert. Openclaw literally brought them customers at the door.
2. Do everyone royally and get them off their platform - with a strong feeling of dislike or hatred towards Anthropic.
Let's see how 2 goes for them. This is not the space to be treating your clients this way.
Graceful handling from Anthropic
https://x.com/bcherny/status/2040206440556826908?s=20
When I do use AI, I already have a solid plan of what I need. Sometimes I ask it to look something up. I never do both in one prompt.
GLM 5.1 can do both, and its way way cheaper. I also don't hit my limit that fast (Plus I get to use it in OpenCode).
I'm hoping that they won't bother you unless you specifically max out the subscription limits every time
Claude Code seems designed to terminate quickly- mine always finds excuses to declare victory prematurely given a task that should take hours.
> To make the transition easier, we’re offering a one-time credit for extra usage equal to your monthly subscription price. Redeem your credit by April 17. We’re also introducing discounts when you pre-purchase bundles of extra usage (up to 30%).
The Anthropic casino wants you to continue gambling tokens at their casino only on their machines (Claude Code) only by giving more promotional offers such as free spins, $20 bets and more free tokens at the roulette wheels and slot machines.
But you cannot repurpose your subscription on other slot machines that are not owned by Anthropic and if you want it badly, they charge you more for those credits.
The house (Anthropic) always wins.
Im hitting rate limits within 1:45 during afternoons.
I can’t justify extra usage since it’s a variable cost, but I can justify a higher subscription tier.
Thanks openclaw for getting me ahead, I’ve taken that and am in Claude code again.
Anthropic's current business model is to sell access to their tools to subscribers at a loss. Users maxing out their $200/month plan can realistically cost Anthropic $500-600 in actual compute costs.
Anthropic is okay with this right now because they want to amass as many users as they can, and eventually hope that GPUs will increase in power and efficiency, and their LLMs will become more efficient as well. They can eventually profit off of their current pricing, or with modest price increases, if that comes to fruition.
But letting OpenClaw wake up every 30 minutes and start sending requests is a surefire way to max out your weekly limits, and that certainly isn't something Anthropic planned for.
I'm doing a side-by-side with GPT-5.4 for $20/mo and Sonnet for $20/mo and I can tell you that all my 5 hour tokens are eaten in 30 minutes with Claude. I still haven't used my tokens for OpenAI.
Code quality seems fine on both. Building an app in Go