The hit open source autonomous AI agent OpenClaw has probably been attracted by Anthropic.
Today, Anthropic announced Cloud Code Channels, a way to connect its own powerful Cloud Code AI agentic harnesses to a human user’s Discord or Telegram messaging application, letting them send direct messages to Cloud Code whenever they want while on the go, and instructing them to write code for them. The official document is here.
It’s not just a new UI; Moving beyond synchronous, it’s a fundamental shift in how developers interact with AI agents "ask and wait" An asynchronous, autonomous partnership model. Previously, cloud code users were stuck interacting with agentic harnesses on cloud desktop applications, terminal or supported developer environments, and cloud mobile apps through a somewhat layered (in my experience) interconnection setting called remote control.
Now, Anthropic is offering some of the same core functionality as OpenClave, which saw its rapid adoption among software developers and Vibe coders following its release in November 2025 by Austrian developer Peter Steinberger (who, ironically, originally called his project Vibe.com). "claws" In honor of Anthropic’s own AI model Cloud, who initially powered it, until Anthropic sent him down for potential trademark violations. Steinberger was hired by Anthropic’s rival OpenAI.)
Central to OpenClave’s appeal was the ability to allow users to have a persistent, personal AI worker, allowing them to send messages 24/7, whenever they felt like it, over general messaging apps like iMessage, Slack, Telegram, WhatsApp, and Discord, and have their AI message them back – not just to chat, but to build entire applications, from composing, sending, and organizing emails and files, to applying for jobs on the user’s behalf, to the entire There is real work to be done for them to manage social marketing campaigns. When AI completes a task, it can immediately alert the human user on their favorite messaging platform.
But OpenClaw also came with a high level of security risk (as it could be given access to a user’s hard drive and file system, or other personal information, and run undetected) and difficulty for non-technical users, prompting a wave of branches promising greater ease and security, including NanoClaw, KiloClaw, and Nvidia’s recently announced NemoClaw.
By giving Cloud Code the same basic functionality – the ability for users to send it messages from the popular third-party apps Discord and Telegram, and it messages them back when it completes a task – Anthropic has effectively countered the appeal of OpenClaw and offered something it doesn’t: the Anthropic brand name with its commitment to AI safety and security, and ease of use right out of the box for less technically inclined users.
Technology: Bridge of Model Reference Protocols
At the heart of this update is the Model Context Protocol (MCP) open source standard that Anthropic introduced in 2024. Think of MCP as a universal USB-C port for AI: It provides a standardized way for AI models to connect to external data and tools. in new "channel" Architecture,An MCP server acts as a two-way bridge.
When a developer starts a Cloud Code session --channels Flag, they’re not just opening chat; They are creating a polling service.
Using the Bun runtime – which is known for its extreme speed in executing JavaScript – Cloud Code monitors specific plugins (currently Telegram and Discord).
When a message arrives, it is injected directly into the active session. <channel> events. The cloud can then use its internal tools to execute code, run tests or fix bugs, and respond back to the external platform using a special reply tool.
The technical achievement here is solidity. Unlike standard web-chat, cloud code sessions can now run in a background terminal or persistent server (like a VPS), waiting for "Humming" Come into action.
How to Set Up Cloud Code Connectors on Telegram and Discord
Setting up these native connectors requires Cloud Code v2.1.80 or later and the BUN runtime installed on your desktop PC or Mac. Follow the instructions here or below.
1. Installation of Telegram
- Create your bot: Open and use BotFather in Telegram
/newbotOrder to generate a unique bot and access token. -
Install plugin: Inside your Cloud Code terminal, run:
/plugin install telegram@claude-plugins-official -
Configure token: run
/telegram:configure <your-token>To save your reputation. -
Restart the channel with: Quit the cloud and restart using the channel flag:
claude --channels plugin:telegram@claude-plugins-official -
Add your account: DM your new bot on Telegram to receive the pairing code, then enter it into your terminal:
/telegram:access pair <code>
2. To establish discord
-
Create an application: Go to the Discord developer portal, create a "new application," And reset the bot token to copy it.
-
Enable Intents: In bot settings, you Sure Able message content intent under "Privileged gateway intent."
-
Install and configure: In Cloud Code, run
/plugin install discord@claude-plugins-officialAfter/discord:configure <your-token>. -
Launch and pair: restart from
claude --channels plugin:discord@claude-plugins-official. DM your bot on Discord and use it/discord:access pair <code>Order to end the link.
Products: Desktop to "everywhere"
The immediate practical impact is the democratization of mobile AI coding. Previously, if a developer wanted to check build status or run a quick fix while away from their desk, they had to rely on complex self-hosted setups like OpenClause.
With channels, the setup is basic. A developer can create a Telegram bot through Botfather, link it to Cloud Code /telegram:configure order, and "couple" Their account with a security code. Once configured, the phone becomes the remote control for the development environment.
The product also gives an introduction "fakechat" Demo—a local-only chat UI that allows developers to test "to push" Logic on your machine before connecting to an external server. This reflects Anthropic’s caution, "Research Preview" approach, ensuring that developers understand the flow of events before exposing their terminal to the Internet.
Licensing: ownership power over open standards
The licensing implications of this release highlight a growing trend in the AI industry: proprietary engines are moving onto the open track. Cloud Code remains a proprietary product tied to Anthropic’s commercial subscriptions (Pro, Max, and Enterprise).
However, by building on the open-source Model Context Protocol, Anthropic is encouraging the creation of a developer ecosystem "connectors" Which makes their model more useful.
While Core Cloud "Brain" Closed, plugins for Telegram and Discord are being hosted on GitHub under the official Anthropic repository, potentially allowing community contributions or forks.
This strategy allows Anthropic to maintain the security and quality of the model while benefiting from the rapid innovation of the open-source community – a straightforward challenge. "Free" But that’s often the fragmented nature of purely open-source agent frameworks.
And because it’s built on MCP, the community can now build "connectors" to Slack or WhatsApp yourself, rather than waiting for Anthropic to send them.
Community Reactions: ‘The OpenClaw Killer’
The reaction from users, especially AI observers on X, was swift and definite. This sentiment was best captured by Ijaz (@CryptoPunk7213), who noted that Anthropic’s speed of shipping – which included texting, thousands of MCP skills, and autonomous bug-fixing in just four weeks – was "Absolutely crazy."
For many, this update makes the local-first agent framework obsolete. BentoBoi (@BentoBoiNFT) observed, "Cloud killed OpenClaw with this update. Now you don’t need to buy a Mac Mini. I say this as someone who has one, hahaha," Referring to the common practice of developers purchasing dedicated hardware to run open-source agents such as OpenClaw 24/7. By taking this persistence to the cloud code environment, Anthropic has simplified "hardware tax" For autonomy.
AI YouTuber Matthew Berman summarized the change: "He created OpenCL."
The consensus among early adopters is that Anthropic has successfully internalized the most desirable features of the open-source movement—multi-channel support and long-term memory—while maintaining the credibility of a tier-one AI provider.
While Anthropic’s cloud has long been a favorite for its logic, it remains "brain in a jar"-A stateless entity that waited for the user to type before he or she could think. Meanwhile, open-source projects like OpenClaw have flourished due to offering "always on" Persistence allows developers to send messages to their AI from Telegram or Discord to trigger complex workflows.
Now, with Anthropic bridging the gap, it’s up to users to decide which approach is best for them.
<a href