Cryptographic Agility Framework for Post-Quantum Secure API Migration: Empirical Performance Analysis and Architectural Blueprint
Year: 2026
Type: article
Abstract: The public-key infrastructure is at risk because quantum computers will soon be useful for encryption. Still, there are a lot of things we don't fully understand about how Post-quantum cryptography (PQC) affects API ecosystems as a whole. This study is the first to do a thorough, organized evaluation of NIST-standard PQC algorithms used in a real-world microservices framework with full API security protocols (TLS 1.3, JWT, and OAuth 2.0). Previous studies only looked at isolated primitives or evaluations that only used TLS. A controlled testing platform shows a big difference in performance: PQC key encoding (Ky-ber768) only causes an 8.2% rise in latency, but Postquantum digital signatures (Dilithium3) make the TLS handshake take 193.6% longer and slow down API traffic by 29.2%. FALCON-512, an alternative lattice-based signature scheme, reduces handshake latency by 18.3% compared to Dilithium3 (108.4 ms vs. 132.7 ms), demonstrating that algorithm selection significantly impacts system-level performance even at equivalent NIST security levels. To deal with these problems, a cryptographic agility framework is suggested and shown to cut migration work by 87.6%, or from 14.5 person-hours to 1.8, which is a real benefit that wasn't seen in previous theoretical approaches. It is also tried out a combined PQC-classical shift approach. It works, but it costs more to run (185% for key management and 120% for log processing) and has a 266.8% handshake overhead. These findings are synthesized into six actionable design principles for Post-quantum API security. This work transforms the PQC transition from a theoretical problem into an evidence-based systems engineering roadmap, enabling organizations to make quantified risk-performance trade-offs when migrating API infrastructure.
Author: Gabriel Silva-Atencio
Institution: Universidad Latinoamericana de Ciencia y Tecnología
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7225405425