Abstract
The present disclosure provides a computer-implemented method for agentic AI-driven vulnerability remediation. A vulnerability (310) is detected and context information associated with the vulnerability is collected. The vulnerability is prioritized and a remediation strategy is selected using a reinforcement learning policy. A patch (320) is generated according to the selected remediation strategy and is associated with a proof bundle (330) comprising machine-checkable evidence including closure evidence (331), regression evidence (332), invariant evidence (333), policy evidence (334), and provenance evidence (335). The patch is validated (340) using the machine-checkable evidence. A quantum-secure provenance record is generated using hash-chained records and post-quantum digital signatures. Governance policies are applied to determine whether remediation is approved. A vulnerability closure decision is generated only when patch validation, governance approval, and provenance verification requirements are satisfied.
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Dubey, Tarushi and Mishra, Shashwat, "AGENTIC PROOF-CARRYING PATCH FRAMEWORK FOR SCALABLE VULNERABILITY REMEDIATION", Technical Disclosure Commons, (November 20, 2026)
https://www.tdcommons.org/dpubs_series/13192