prolog-reasonerAI & Machine LearningCode ExecutionAutonomous AgentsrikarazomeAlicenseAqualityDmaintenanceSWI-Prolog execution for LLMs with CLP(FD) and recursion — boosts logic/constraint accuracy from 73% to 90% on a 30-problem benchmark. Updated 5 months ago (2026-05-01 20:51 UTC)511MIT