ProsaBuddy: Assisting Mechanized Real-Time Schedulability Analysis with LLM-based Agents

2026-10-01

AI Summary

This study addresses the high expertise barrier, error-proneness, and labor intensity of manually constructing formal proofs for real-time scheduling. We propose a large language model-based multi-agent automated proving system that integrates ReAct reasoning loops with a subgoal delegation architecture. By leveraging retrieval-augmented generation to invoke code libraries, the system drives the Rocq proof assistant to perform automated theorem proving while supporting optional human-in-the-loop feedback. Experimental results demonstrate that our approach significantly outperforms existing Rocq automation tactics and the general-purpose coding agent OpenCode on benchmark tasks. This work effectively reduces the proof overhead associated with machine-verifiable scheduling, offering an efficient, intelligence-driven paradigm for the formal verification of high-assurance real-time systems.

Abstract

Rigorous schedulability analysis is essential for the design of hard real-time systems, yet errors in pen-and-paper proofs threaten the safety of critical applications. The Prosa initiative addresses this by offering a foundation for building machine-checkable schedulability analysis proofs in the Rocq proof assistant. However, the substantial time and expertise required to construct such proofs remain a major barrier for wider adoption of Prosa. This work presents ProsaBuddy, an LLM?based agent system designed to lower the effort needed to develop mechanized real-time schedulability proofs. ProsaBuddy employs a ReAct loop with retrieval over the Prosa codebase, access to Rocq tools and optional human-written hints. It uses a subgoal?delegation architecture, decomposing a lemma into subgoals and dispatches them to subagents for proof. We evaluate ProsaBuddy on a mini benchmark drawn from real-time scheduling literature. Experiment results show that ProsaBuddy significantly outper?forms state-of-the-art LLM-based Rocq automated proving agent systems and a general coding agent OpenCode

Problem

• real-time systems
• schedulability analysis
• mechanized proofs
• proof assistant
• formal verification

Innovation

• LLM-based Agents
• Mechanized Schedulability Analysis
• ReAct Loop
• Subgoal-Delegation Architecture
• Rocq Proof Assistant