World Hashes demonstrates three national cryptographic hash standards — SM3 (China, OSCCA, 2010), Streebog (Russia, FSB, 2012), and Kupyna (Ukraine, 2014) — alongside SHA-256 and SHA-3 as reference anchors. Each national hash was developed for cryptographic sovereignty: to reduce dependence on U.S.-designed primitives for government and regulated-industry use. The security model is collision-resistant one-way function: given a hash output, finding the input or a second input with the same hash must be computationally infeasible.

The page opens with a ground-floor "what is a cryptographic hash?" panel that defines the four properties — deterministic, fixed-length, one-way, collision-resistant — and demonstrates them live (type anything and watch the digest stay 256 bits while a single extra character scrambles it). Every construction term (Merkle–Damgård, sponge, wide-pipe, S-box, Miyaguchi–Preneel, Keccak-f[1600], length-extension) is glossed inline on hover/focus, and a "Constructions 101" note contrasts block-chaining against absorb-and-squeeze so a newcomer can follow the rest.

• SM3: Required for Chinese PKI, Chinese TLS (TLCP), and products under Chinese Cryptography Law — pairs with SM2 signatures.
• Streebog-256/512: Required for Russian GOST R 34.11-2012 compliance — pairs with GOST elliptic curve signatures.
• Kupyna-256/512: Required for Ukrainian DSTU 7564:2014 compliance.
• SHA-256: General-purpose default for all other use cases.
• SHA-3: Preferred for new protocol designs or when sponge security matters.
• Do not use Streebog outside Russian compliance requirements — shared S-box transparency concerns with Kuznyechik apply.
• Do not use any of these as a MAC without HMAC wrapping.
• Do NOT treat this as a production crypto library — it is a teaching demo for comparing national hash standards, not a hardened deployment.

systemslibrarian.github.io/crypto-lab-world-hashes

A ground-floor intro panel plus five exhibits:

1. What is a cryptographic hash? — the four properties in plain language, demonstrated live: type anything and watch the SHA-256 digest stay a fixed 256 bits while one extra character scrambles it; a collapsible explains one-wayness and collisions.
2. SM3 (China) with SHA-256 side-by-side, a labelled Merkle–Damgård mechanism diagram (message blocks chaining through a compression function to an exposed final state → length-extension), and a user-driven avalanche — click any character to flip it yourself and watch the digest re-scramble.
3. Streebog (Russia) with S-box controversy documentation and its wide-pipe MD construction glossed.
4. Kupyna (Ukraine) with geopolitical context and a sponge mechanism diagram (absorb into a wide rate/capacity state, squeeze out the digest, capacity never exposed → no length-extension).
5. SHA-256 and SHA-3 as reference anchors with five-way simultaneous hashing.
6. Five-way comparison table + decision tree, with the Trust-level column explicitly labelled editorial opinion and each grade sourced in an expandable note (e.g. Streebog's caution cites Perrin et al. 2019 on the shared S-box).

All hash outputs are real — no simulation. Every digest is produced in your browser by src/hashes.ts, the same module the test suite checks. Each exhibit shows a visual avalanche diff (changed hex nibbles highlighted, with bit-diffusion percentage against the ideal ~50%) and a live input byte-length readout. Nothing is sent to a server — all hashing is local.

• Using a raw Merkle–Damgård hash (SM3, SHA-256, Streebog) directly as a MAC is vulnerable to length-extension; wrap it in HMAC instead.
• Streebog and Kuznyechik share S-box transparency concerns, so deploying Streebog outside its compliance mandate inherits an unresolved design-trust question.
• Reaching for a national hash outside its regulatory requirement trades SHA-2/SHA-3 scrutiny and tooling for weaker ecosystem support with no security gain.
• Collision and second-preimage resistance are assumptions, not guarantees; truncating a digest or misusing it for password storage (no salt/KDF) undermines the intended security.
• Mismatched variants (e.g. confusing the 256-bit and 512-bit outputs, or differing byte/endianness conventions between standards) break interoperability between implementations.

• SM3 underpins Chinese PKI, Chinese TLS (TLCP), and products under Chinese Cryptography Law, paired with SM2 signatures.
• Streebog-256/512 is mandated for Russian GOST R 34.11-2012 compliance, paired with GOST elliptic-curve signatures.
• Kupyna-256/512 is required for Ukrainian DSTU 7564:2014 compliance.
• SHA-256 (FIPS 180-4) remains the general-purpose default across the internet, with SHA-3 (FIPS 202) preferred for new sponge-based designs.

git clone https://github.com/systemslibrarian/crypto-lab-world-hashes
cd crypto-lab-world-hashes
npm install
npm run dev

• crypto-lab-world-ciphers — the encryption counterpart: Camellia, ARIA, SM4, and Kuznyechik national block ciphers.
• crypto-lab-hash-zoo — SHA-256, SHA3-256, BLAKE3, and the Merkle–Damgård construction explained.
• crypto-lab-babel-hash — interactive SHA-256, SHA3-256, BLAKE3, and HMAC.
• crypto-lab-mac-race — HMAC, CMAC, Poly1305, and GHASH: turning hashes into authenticators.
• crypto-lab-merkle-vault — Merkle trees and inclusion proofs built on a hash function.

The hero shows a live self-test badge (e.g. ✓ 17/17 test vectors verified). On every page load the app recomputes published known-answer test vectors and compares them byte-for-byte against the values in each algorithm's defining standard:

The same vectors run in CI via Vitest on every push, alongside a happy-dom render test that asserts the UI mounts and reports all vectors passing, and an axe-core accessibility scan.

npm test         # known-answer + render + accessibility tests
npm run test:watch
npm run typecheck

The lab is built to WCAG 2.1 AA: semantic landmarks and a skip link, the full ARIA tab pattern with arrow/Home/End keyboard navigation, visible focus rings, focus and caret preserved across re-renders, scope-annotated data tables, prefers-reduced-motion support, and a <noscript> fallback. Avalanche highlights carry three independent cues (colour, background, underline) so they never rely on colour alone, and colour contrast meets AA in both themes.

npm test runs an automated axe-core scan (WCAG 2 A/AA) against the rendered app in both themes on every push. Layout-dependent colour-contrast is verified separately, since a headless DOM has no rendering engine.

This project is configured for automatic Pages deploy from GitHub Actions.

1. Push this repo to GitHub on the main branch.
2. In GitHub: Settings -> Pages -> Build and deployment.
3. Set Source to GitHub Actions.
4. Push to main (or run the Deploy to GitHub Pages workflow manually).

Manual deploy is also available:

npm run deploy

"So whether you eat or drink or whatever you do, do it all for the glory of God." — 1 Corinthians 10:31