mlsanityData to 7 Oct 2026

Agentic Method for Deterministic Validation of Legacy Code Migration

arXiv 2607.28271 ↗First posted 30 Jul 2026cs.SE cs.AIcited by 0

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

Migration of legacy COBOL programs to Java requires extensive testing to ensure correct functionality. This effort is often complicated by the lack of test data and the difficulty of validating all corner cases. In this paper we propose a novel agentic test-synthesis method, the "Locksmith Loop," which is initiated by preparing two runtime environments: the COBOL source and the generated Java target are each instrumented with mocks and executed off-mainframe on commodity hardware, then an iterative agentic loop performs Witness Search over input mocks to penetrate program branches, followed by parity-preserving mutations. When routing boundaries are reached, an analyzer identifies a Locked Paragraph: a condition preventing deeper exploration. Across three COBOL-Java case studies, spanning two open-source programs and one internal production-like COBOL program and ranging from 430 to 4,114 source lines, Locksmith consistently improved coverage beyond input-search plateaus, reaching nearly complete coverage on the two open-source programs and 91.90% branch coverage on the internal production-like COBOL program. The generated Java matched the COBOL reference under deterministic parity checks in all accepted test cases. Through these findings we demonstrate, to the best of our knowledge, a novel approach for validating agentic coding output using a deterministic oracle.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 22 of its 37 references that are in the map. First posted on arXiv 30 Jul 2026.

Builds on

22 of 37 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering6 May
  2. 2024RepairAgent: An Autonomous, LLM-Based Agent for Program Repair25 Mar

Other references · 20

  1. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
  2. 2015Illuminating search spaces by mapping elitescited by 406
  3. 2023An Empirical Evaluation of Using Large Language Models for Automated Unit Test Generationcited by 193
  4. 2024Automated Unit Test Improvement using Large Language Models at Metacited by 73
  5. 2018The Art, Science, and Engineering of Fuzzing: A Surveycited by 61
  6. 2023A Survey of Learning-based Automated Program Repaircited by 55
  7. 2016A Survey of Symbolic Execution Techniquescited by 36
  8. 2025A Survey of LLM-based Automated Program Repair: Taxonomies, Design Paradigms, and Applicationscited by 24
  9. 2021Monte Carlo Elites: Quality-Diversity Selection as a Multi-Armed Bandit Problemcited by 10
  10. 2021An Empirical Study of OSS-Fuzz Bugscited by 7
  11. 2018A survey on software testabilitycited by 6
  12. 2024Mokav: Execution-driven Differential Testing with LLMscited by 6
8 more
  1. 2024XMainframe: A Large Language Model for Mainframe Modernizationcited by 6
  2. 2024DiffSpec: Differential Testing with LLMs using Natural Language Specifications and Code Artifactscited by 5
  3. 2025Automated Testing of COBOL to Java Transformationcited by 4
  4. 2017A Snowballing Literature Study on Test Amplificationcited by 3
  5. 2025Mut4All: Fuzzing Compilers via LLM-Synthesized Mutators Learned from Bug Reportscited by 3
  6. 2025Automated Validation of COBOL to Java Transformationcited by 2
  7. 2018Automatic Test Improvement with DSpot: a Study with Ten Mature Open-Source Projectscited by 1
  8. 2026COBOL-Coder: Domain-Adapted Large Language Models for COBOL Code Generation and Translationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Dynamic Cogeneration of Bug Reproduction Test in Agentic Program Repair
  2. 2026SWE Refactor Bench: Can Coding Agents Complete a Long-Horizon, Whole-Repository Stack Migration?
  3. 2025FreshBrew: A Benchmark for Evaluating AI Agents on Java Code Migration
  4. 2026Deterministic vs. LLM-Controlled Orchestration for COBOL-to-Python Modernization ↗
  5. 2026TestJack: Should you trust the results in coding benchmarks? Agentic Coding Benchmarks Auditing via Evaluator Evolution
  6. 2025QualityFlow: An Agentic Workflow for Program Synthesis Controlled by LLM Quality Checks
  7. 2026Breaking, Stale, or Missing? Benchmarking Coding Agents on Project-Level Test Evolution
  8. 2025Nexus: Execution-Grounded Multi-Agent Test Oracle Synthesis