Clawdbot started as a “personal AI agent” experiment by developer Peter Steinberger—something closer to a doer than a chatter. Instead of answering questions, it aimed to complete tasks across real services: messaging, email, calendars, and more, driven by an LLM and a growing set of integrations.
The early hook was simple: run your own assistant and connect it to the channels you already live in—WhatsApp, Telegram, Slack, Discord, iMessage/Teams bridges, and similar connectors—so it could operate where work already happens.
Clawdbot’s name deliberately riffed on Anthropic’s “Claude” ecosystem—especially the Claude Code branding and its “Clawd” mascot—creating instant recognition but also obvious confusion about affiliation. That attention helped the project spread extremely fast.
Within weeks, it was being described as a breakout open-source agent project, with traction numbers that turned it into a mainstream AI story rather than a niche GitHub repo.
