Clawing at OpenClaw: The Claude Code Channels Controversy
Key Takeaways:
-
In January 2026, Anthropic filed a trademark complaint, forcing OpenClaw to rebrand from Clawdbot.
-
In February 2026, Anthropic shut down a billing loophole, banning OAuth token use in third-party tools.
-
The ban forced OpenClaw users to transition to pay-per-token API keys, skyrocketing their operating costs.
-
Weeks after the ban landed, Anthropic released Claude Code Channels, with the same functionalities as OpenClaw, but paid.
-
The sequence of events made many believe this was deliberate competitor elimination, not policy enforcement.
Executive Summary: Anthropic closed the exploitative OAuth billing loophole, effectively banning OpenClaw from Claude. Only weeks after the ban landed, they released Claude Code Channels with replicated OpenClaw features and immediately monetized it. The timeline of events led many to believe this was a calculated move to eliminate a competitor, rather than a justified policy enforcement.
In late 2025, a lone Austrian developer was quietly building something revolutionary: a framework that would elevate Anthropic’s Claude experience by making it persistent and remotely accessible – OpenClaw.
The users loved it. Anthropic didn’t.
What’s the feud between OpenClaw and Anthropic about?
Truth be told, there was never a “feud” in the true sense of the word. Only a series of tactical and controversial moves undertaken by Anthropic, resulting in OpenClaw being crippled as a Claude-centric tool and replaced by Channels – Anthropic’s own framework with virtually identical functionality.
How it all started
Apparently, OpenClaw has been a thorn in Anhtropic’s side for a while now, and it all began in early 2026, when OpenClaw was still called Clawdbot. Deeming Clawd too similar to Claude, Anthropic’s legal team asked Steinberger to change the name, which he did.
🦞 BIG NEWS: We've molted!
— OpenClaw🦞 (@openclaw) January 27, 2026
Clawdbot → Moltbot
Clawd → Molty
Same lobster soul, new shell. Anthropic asked us to change our name (trademark stuff), and honestly? "Molt" fits perfectly – it's what lobsters do to grow.
New handle: @moltbot
Same mission: AI that actually does…
Though legally grounded, this move was seen as petty by many, especially given that ClawdBot was driving millions of users to Claude on a virtually daily basis. Well, as it turns out, that actually was the problem.
The crackdown
The real conflict centers around Anthropic shutting down on a major billing loophole exploited by OpenClaw users. In late 2025, users discovered they could bypass Claude’s API pricing by extracting an OAuth token from their Pro/Max subscription, effectively allowing them to route massive workloads through flat-rate $200 subscription plans.
However, since OpenClaw is so incredibly resource-hungry, it was burning through API usage orders of magnitude higher than subscription costs. Naturally, this created a completely unsustainable cost asymmetry, causing Anthropic to respond by tightening its policies and banning OAuth token usage in third-party tools.
The controversy
Although Anthropic didn’t outright ban OpenClaw – just the subsidy exploit – it effectively forced the transition to pay-per-token API keys. Consequently, users were suddenly facing skyrocketing operating costs, leading to a widespread disruption and user “meltdown.”
However, despite the backlash, many users actually saw Anthropic’s move as justified – a company protecting its resources and preventing exploitation. And then they went and released the exact same tool…
At this point, it’s like Anthropic’s taking notes straight from Google’s playbook: take something somebody else built, sanitize it, cannibalize it – and monetize it. On March 20, 2026 – just weeks after formalizing the ban on OAuth token usage – Anthropic released a massive update called Claude Code Channels.
Powered by Anthropic’s Model Context Protocol (MCP) and Bun runtime, this upgrade essentially turned Claude into an ambient assistant through features such as:
- Messaging app integration – allows users to interact with Claude from anywhere using Discord and Telegram.
- Background persistence – enabling the Claude Code session to run in a background terminal, and spring into action when users ping it.
Sounds familiar? It should. That’s 1:1 what OpenClaw did – give Claude persistent memory and allow users to manage their local machines remotely – for free. That’s right – Anthropic didn’t just release the same functionality, but monetized it promptly. Now, it all comes together.
We just released Claude Code channels, which allows you to control your Claude Code session through select MCPs, starting with Telegram and Discord.
— Thariq (@trq212) March 19, 2026
Use this to message Claude Code directly from your phone. pic.twitter.com/sl3BP2BEzS
Did Anthropic deliberately kill OpenClaw?
All the evidence points to “yes.” The trademark complaint, the banning of OAuth under the guise of “rule enforcement,” the release of the exact same workflow, locking the users into Anthropics pricing system – it’s textbook sherlocking.
In one fell swoop, Anthropic eliminated its fastest-growing open-source competitor, absorbed its ideas, and capitalized on the market it established. Some would say this was a brilliant, albeit ruthless, business maneuver. The users saw it as a heavy-handed, self-serving move, and the ultimate act of hypocrisy.
Anthropic – a company that positioned itself as the paragon of ethics, a pioneer of principled AI, a shiny beacon of light among power and money-hungry competitors – resorting to tactics many were convinced was beneath them. The question now is whether this will “slide” – or can we expect users to give Claude the same treatment they did ChatGPT?
AI power-grab is in full swing
If you’re waiting for the dust to settle – you’re already losing.
Anthropic may have just pulled the oldest move in Big Tech’s playbook, but they are not the only ones rewriting the rules.
AI is reshaping the search and the market as a whole – and it’s doing it in real time.
The burning question is, will you act now, or will you contend with whatever remains, if anything?
If you’re in the former category, you’re in the right place.
ZeroClick Labs offers full-scope SEO for AI search engines, with systems designed to anticipate algorithm shifts rather than chase them.
“Our agency had no idea how to approach AI visibility. ZeroClick only does this one thing so they actually know what works. Worth every penny just to not waste time figuring it out ourselves.” – Jay