Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{195018,
        author = {Mihir Patil and Pranav Patil and Siddhesh Patil and Prerana Mhatre and Shailesh Galande},
        title = {AutoCure: AST-Grounded, Confidence-Gated Self-Healing for Runtime Errors and GitHub Code Review},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {5618-5626},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195018},
        abstract = {This paper describes results-focused research on an AI-assisted self-healing software system named AutoCure that incorporates runtime error analysis and GitHub code review in a single system. The implemented system includes a variety of features such as production log ingestion via WebSocket technology, language-based error parsing, source path normalization, abstract syntax tree contextual tracing, confidence validation, and policy-gated remediation actions. The review process also enhances traditional diff analysis by comparing new and old abstract syntax tree structures, tracing changed symbol reference tracing, and generating review flags for low utility changes or redundant code. The orchestration is implemented as a FastAPI application with components such as log analysis, abstract syntax tree services, AI services, confidence scoring, GitHub services, email generation, and report persistence. The study demonstrates that AST evidence plus confidence gating can convert AI outputs into bounded, auditable engineering actions.},
        keywords = {AST analysis, code review automation, confidence validation, production log ingestion, self-healing software.},
        month = {March},
        }