QCrypton: A Unified Platform for AI/LLM Threat Detection and Post-Quantum Cryptographic Security Assessment

The convergence of large language model (LLM) deployments and the approaching quantum computing era creates unprecedented security challenges that no single existing tool addresses holistically. We present QCrypton, a unified security platform that integrates an 11-scanner AI threat detection engine with a post-quantum cryptographic assessment framework, fault-tolerant quantum computer (FTQC) attack cost estimation, and a zero-native-dependency cryptographic toolkit. QCrypton bridges the gap between AI security and quantum readiness by providing bidirectional integration: threat scanners invoke quantum vulnerability assessments in real time, while detected secrets are encrypted via BB84 quantum key distribution simulation before reporting. We describe the system architecture, evaluate scanner accuracy across 22+ prompt injection patterns and 8 MCP tool poisoning vectors, and demonstrate the FTQC cost estimation engine across 28 cryptographic algorithms. QCrypton is implemented in Node.js with zero native cryptographic dependencies, using only platform-native OpenSSL primitives, eliminating supply-chain risk in the cryptographic layer.