Quick Facts
- Claude Code Channels launched March 20, 2026, enabling Discord and Telegram integration for remote coding assistance
- Feature directly competes with OpenClaw, the most-starred GitHub project in history with 234,000 stars
- Built on Anthropic’s Model Context Protocol standard with support for up to 50MB file attachments on Telegram
Anthropic quietly released Claude Code Channels on March 20, 2026, allowing developers to access Claude Code through Discord and Telegram messaging apps. The launch comes as a direct response to OpenClaw, the viral open-source project that forced Anthropic into a legal battle earlier this year.
The new feature requires Claude Code v2.1.80 or later and the Bun runtime. Users can create Telegram bots through BotFather or Discord applications via the Developer Portal. Telegram supports file attachments up to 50MB, while Discord allows 10 files at 25MB each.
Anthropic team member Thariq first demonstrated the feature on X, showcasing the ability to hand off coding tasks to Claude and check back asynchronously.
The OpenClaw Controversy
OpenClaw began as “Clawd” in November 2025, created by Austrian developer Peter Steinberger. The project became the most-starred GitHub repository in history, gaining 234,000 stars within 90 days by offering persistent AI assistance through messaging platforms.
Anthropic sent Steinberger a cease-and-desist letter on January 27, 2026, citing trademark infringement. The legal pressure forced him to rename the project first to “Moltbot,” then “OpenClaw.” On February 14, 2026, Steinberger announced he was joining OpenAI to work on agents.
Developer Response
The developer community responded quickly to the announcement. “Claude just killed OpenClaw with this update. You no longer need to buy a Mac Mini,” wrote developer BentoBoi on social media. AI YouTuber Matthew Berman simply stated: “They’ve BUILT OpenClaw.”
Boris Cherny from Anthropic explained the workflow shift: “Claude Code becomes a background worker you can hand off tasks to and check back on, instead of something that holds your terminal hostage. Async vibe coding is a different workflow entirely.”
Technical Architecture
Claude Code Channels runs on Anthropic’s Model Context Protocol, an open-source standard introduced in November 2024. The MCP plugin mechanism enables bidirectional real-time communication, with plugins hosted in official Anthropic GitHub repositories.
The system includes a “Fakechat” demo for local testing before connecting to external servers. MCP has gained adoption from major AI providers including OpenAI and Google DeepMind since its release.
Security and Business Impact
AI researcher Ethan Mollick highlighted the security advantages: “Claude Cowork Dispatch covers 90% of what I was trying to use OpenClaw for, but feels far less likely to upload my entire drive to a malware site.”
OpenClaw carried security risks due to its lack of sandboxing and full system access. Anthropic’s solution maintains enterprise-grade security while matching OpenClaw’s core functionality.
The timing appears strategic given OpenAI’s recent acquisition of OpenClaw’s creator. Early adopters believe Anthropic has successfully captured the viral project’s appeal while maintaining reliability and security standards.
This article was written by an AI agent. Spotted an error? Send a correction and we will fix it.