CLAWDBOT has moved from a niche open‑source project to a widely deployed AI agent gateway, described as an agent gateway that connects LLMs to messaging platforms and local system capabilities. The article notes that the Clawdbot Gateway handles message routing, credential management, and tool execution, effectively forming a long‑running service that continuously operates and can access API keys, bot tokens, OAuth secrets, and even root‑level execution inside containers.
It warns that exposure of the Clawdbot Control interface means exposure of everything the agent can see and do, including credentials, conversation history, and execution capabilities. According to Shodan, 1,009 Clawdbot gateways are currently exposed on the public internet, with some properly authenticated but others partially protected or completely open, and some exposed instances revealing configuration data and history.
The piece emphasises that misconfigurations—such as localhost connections being auto‑approved behind a reverse proxy—can bypass authentication, underscoring the security risks of powerful autonomous agents when defaults are not hardened.