Clawdbot Rebrands to Moltbot Following Anthropic Trademark Request Amidst Viral Growth and Security Warnings

The popular open-source AI assistant, Clawdbot, has rebranded to Moltbot after a trademark request from Anthropic, facing both rapid adoption and serious security concerns.

The rapidly rising open-source AI assistant, initially known as Clawdbot, has officially rebranded to Moltbot, effective January 27, 2026. This change, which also saw its companion AI named “Clawd” become “Molty,” was prompted by a trademark request from AI developer Anthropic, citing similarities with their product, Claude, and mascot, Clawd.

Created by developer and entrepreneur Peter Steinberger, Clawdbot quickly gained significant traction, accumulating over 60,000 GitHub stars within weeks of its November 25, 2025 release. It was lauded for its “agentic AI” capabilities, distinguishing itself from traditional chatbots by focusing on executing tasks rather than merely generating text. The project integrates directly with widely used communication platforms such as WhatsApp, Telegram, Slack, Discord, Google Chat, and Microsoft Teams, functioning as a self-hosted personal AI assistant that operates on users’ own devices.

The rebranding occurred swiftly after Anthropic’s legal team issued a trademark notice in January 2026. Peter Steinberger, Clawdbot’s founder, explained the rationale, stating, “Anthropic asked us to change our name (trademark stuff), and honestly? ‘Molt’ fits perfectly — it’s what lobsters do.” The team embraced a biology-inspired metaphor, likening the change to a lobster molting to grow, asserting that the project’s core mission of “building AI that actually does things, not just chats” remains unchanged.

Despite its innovative approach, the project has also drawn significant security warnings. Experts have cautioned against installing Clawdbot (now Moltbot) without a thorough understanding of the risks. According to Prompt Security CEO Itamar Golan, “a disaster is coming” due to thousands of Clawdbot gateways reportedly exposed on the public internet with no authentication but full shell access. Reports indicate real attacks, including failed login attempts, theft of personal accounts, and prompt injection attacks resulting in data loss. The inherent power of Clawdbot as an autonomous agent capable of executing commands, reading/writing files, and accessing users’ digital lives around the clock underscores these concerns. The project’s architecture includes a “session isolation” mechanism, dividing usage into “main sessions” with full system permissions and “non-main sessions” that run in a Docker container sandbox mode to mitigate risks.

As Moltbot moves forward under its new identity, the team emphasizes that existing users will experience no functional changes, with only branding and handles updated. The project’s official repository is now github.com/moltbot/moltbot, and its documentation is available at docs.molt.bot. However, former Clawdbot addresses have reportedly been maliciously hijacked, with warnings issued against visiting them due to potential security risks from malware or cryptocurrency scams.