Agent Bayes serves as an AI-driven research aide tailored for graduate students, postdocs, and academics alike. Users can upload papers in PDF format or sync their libraries from Zotero, subsequently posing inquiries related to their collection. The assistant meticulously searches through these materials, constructing a mindmap that illustrates claims, each connected to the specific passage and page that substantiates it. Users have the ability to click on a citation to view the relevant passage or PDF page, allowing competing research findings to coexist on the map as distinct sourced viewpoints. Additionally, the AI Verification feature evaluates whether the cited passages adequately support a claim, although this capability is reserved for those on paid plans. An MCP server facilitates the integration of compatible tools like Claude Code, Codex, and Cursor, enabling them to search the user's library and leverage its cited answers and mindmaps. The free plan offers 50 credits each month, while paid subscriptions initiate at $11 per month, billed on an annual basis, providing users with a range of options to enhance their research experience. This innovative assistant aims to streamline the research process and improve academic productivity.