An open-source runtime that gives AI agents unified, controlled access to files, commands, Git, browser automation, and task execution across local machines, remote servers, and containers via MCP.
Adapted from @DanKornasRunning an AI agent across machines shouldn’t mean juggling SSH sessions. AgentDock is an independent tool runtime for AI agents that need to operate local computers, remote servers, and containers. It helps you run multi-machine work from one AI conversation by exposing controlled access to files, commands, Git, browser automation, and task execution through MCP. Key features: • Unified tool access – brings file, command, Git, Skills, MCP, browser automation, and task execution into one runtime • Multi-device coordination – connect multiple AgentDock instances to work across machines in one conversation • Environment coverage – supports local computers, remote servers, and containers • Runtime visibility – inspect logs, processes, ports, and actual runtime state • Recoverable tasks – persist long-running task state and continue after an interruption It’s open-source (Apache License 2.0 license). Link in the reply 👇
