Come learn directly from the AI4RA team! Want to learn more about how to incorporate AI safely and effectively into your sponsored programs office? Learn directly from AI4RA team members at the upcoming 2026 AI Symposium and Annual Meeting hosted by the National Council of University Research Administrators. We’d love to see you there!

MindRouter is an AI gateway developed by Luke Sheneman at the University of Idaho’s Research Computing and Data Services (RCDS). It is freely available for you to install and use under an open-source software license. It sits as a middle layer between tools like Vandalizer and the AI models that power them, routing requests across a cluster of GPU servers and providing a single, reliable access point for running large language models on university-controlled infrastructure.

The Prior Approval Requirements Tracking Workflow reads through agreement documents and produces a comprehensive Prior Approval reference document. It extracts information on activities that require sponsoring agency authorization, including budget-related approvals, scope and timeline changes, and administrative approvals.

The Effort Reporting Compliance workflow analyzes federal award documents and produces a structured Effort Compliance Brief.

The FFR Management Extraction workflow is an AI-powered tool in Vandalizer that analyzes federal award documents and extracts everything you need to manage your Federal Financial Report (FFR/SF-425) obligations. This workflow processes federal award documents, such as award notices, cooperative agreements, or award terms and conditions, and it produces a structured, markdown-formatted reference

The Subaward Agreement Extraction workflow is a Vandalizer post-award tool that reads subaward agreements and pulls relevant data for award setup and monitoring all in one run. It processes subaward agreements, pass-through entity documents, and subaward modifications, handling documents of varying length, including attachments and referenced terms. 

Writing budget justifications is a common time-intensive pre-award task, requiring careful cross-referencing between budget figures, fringe rates, indirect cost calculations, and the project narrative. This workflow automates that process while maintaining accuracy.

The RFA (Request for Applications) Checklist Extraction workflow is a two-step, AI-enhanced process that turns an agency RFA, FOA, or NOFO into a structured pre-award checklist for the proposal team. It runs six parallel extraction tasks against the document — dates and deadlines, eligible institutions, eligible individuals, award information, application components, and budget requirements and policies — then a consolidation step assembles the results into a single Markdown checklist organized in eight sections. Missing information is explicitly marked “Not specified in the document” rather than guessed, and strict de-duplication rules keep each fact in exactly one section. 

What is a workflow and how can it help me automate an RA process? A workflow is an ordered series of tasks which describe a process from start to finish.

The eCFR MCP Server is a free, open-source tool created by Sabrina Woo and Nathan Layman of Artificial Intelligence for Research Administration (AI4RA) that connects AI assistants directly to the Electronic Code of Federal Regulations (eCFR) API.