An adaptive cryptographic library and workbench for high-integrity, profile-driven encryption.
Mango is a next-generation cryptographic engine designed to intelligently adapt to the structure of input data. It uses reversible transformation sequences—optimized per profile—to deliver encryption that’s both robust and tunable.
Whether encrypting random, structured, or natural text data, Mango applies tailored sequences that maximize entropy, diffusion, and key sensitivity.
Mango includes both a C# library and a companion tool, the Workbench, for:
Traditional ciphers apply static logic to every input. Mango takes a different approach: it adapts to the structure of each data set, selecting transform sequences that disrupt patterns and elevate security outcomes.
Mango isn’t just theory — it’s a working encryption system with a modular library, a fully-featured Workbench, and a tunable profile engine.
In this first video, I introduce the architecture and walk through how Mango uses profiles to control encryption behavior. It also covers the relationship between the library and the Workbench, profile creation, and an overview of sequence optimization.
You can watch it here: Watch on YouTube
More videos will follow over time as I dive deeper into specific components and use cases.
Mango is an adaptive encryption engine that tailors its behavior to the structure of the data — fast, tunable, and cryptographically resilient.
This follow-up to Mango’s core architecture dives into the InputProfile — the programmable layer that drives adaptive encryption through transform sequences and round configurations.
Munge is the engine behind Mango’s adaptive cryptography. This internal system explores millions of transform combinations to forge the optimized sequences that drive Mango’s encryption — pushing it to the edge of cryptographic performance.
Mango isn’t just encryption — it’s cryptographic obfuscation. This article explores Mango’s defense against modern threats through adaptive algorithms, data-shaped keying, and metadata-free ciphertext.
Mango’s built-in two-factor encryption goes beyond access control. Both the password and ZoneInfo actively shape the ciphertext itself—mutating the cryptographic output with no metadata or signal. This approach delivers deeper entropy and makes recovery impossible without both secrets.
The Tomasello Signature Vector (TSV) is a deterministic 32-byte fingerprint that reflects what data is — not just its identity. Compact, content-aware, and platform-independent, TSV enables rapid classification without headers, metadata, or machine learning.
Mango’s InputProfiles act like control programs for encryption. Instead of static behavior, profiles tell the cipher how to behave — which transforms to use, in what order, and with how many rounds — adapting cryptographic behavior to the input itself.