Build Your Personalized Research Group: A Multiagent Framework for Continual and Interactive Science Automation

The automation of scientific discovery represents a critical milestone in Artificial Intelligence (AI) research. However, existing agentic systems for science suffer from two fundamental limitations: rigid, pre-programmed workflows that cannot adapt to intermediate findings, and inadequate context management that hinders long-horizon research. We present $\mathtt{\text{freephdlabor}}$, an open-source multiagent framework featuring $\mathit{\text{fully}}\mathit{\text{dynamic}}\mathit{\text{workflows}}$ determined by real-time agent reasoning and a \coloremph{$\mathit{\text{modular}}\mathit{\text{architecture}}$} enabling seamless customization – users can modify, add, or remove agents to address domain-specific requirements. The framework provides comprehensive infrastructure including $\mathit{\text{automatic}}\mathit{\text{context}}\mathit{\text{compaction}}$, $\mathit{\text{workspace-based}}\mathit{\text{communication}}$ to prevent information degradation, $\mathit{\text{memory}}\mathit{\text{persistence}}$ across sessions, and $\mathit{\text{non-blocking}}\mathit{\text{human}}\mathit{\text{intervention}}$ mechanisms. These features collectively transform automated research from isolated, single-run attempts into $\mathit{\text{continual}}\mathit{\text{research}}\mathit{\text{programs}}$ that build systematically on prior explorations and incorporate human feedback. By providing both the architectural principles and practical implementation for building customizable co-scientist systems, this work aims to facilitate broader adoption of automated research across scientific domains, enabling practitioners to deploy interactive multiagent systems that autonomously conduct end-to-end research – from ideation through experimentation to publication-ready manuscripts.