Crypto Schola
What's inside

125 challenges across the TLS attack surface.

Ten families cover client configuration and verification, protocol downgrades, RSA, DH/ECDH, ECDSA, RNG failures, post-quantum cryptography, oracles and key leakage. Most labs ship in both TLS 1.2 and TLS 1.3 variants, plus multi-rounds challenges that mix vulnerabilities.

Client trusts the wrong thing

No chain check · SAN-only match · leaked Superfish CA key

Real client verification bugs

Apple "goto fail" · CVE-2020-0601 forged base point

Protocol downgrade

NULL cipher · FREAK export RSA · Logjam DHE

Broken RSA keys

Fermat · Wiener · Boneh-Durfee · Pollard p±1 · RSA-CRT faults

Weak Diffie-Hellman parameters

Tiny modulus · smooth p-1 · bad generator · invalid curve · MOV

Signing-nonce leakage

Nonce reuse · biased nonce · lattice attack

Post-quantum cryptography

ML-KEM/Kyber · HQC · ML-DSA/Dilithium · SLH-DSA/SPHINCS+ · KEMTLS · nonce reuse · weak keys

Weak or backdoored entropy

Predictable ServerHello.random · weak X25519 keys

Padding & compression oracles

Bleichenbacher · Manger · BREACH / CRIME

Key leakage & side channels

FTP · SSH · HTTP · Redis · Postgres · Heartbleed

Proof, not theory

Examples of real attacks reproduced.

Each family above isn't hypothetical — these are named CVEs and historical incidents the lab rebuilds at the protocol level.

CVE-2014-0160 Heartbleed
CVE-2014-1266 Apple "goto fail"
CVE-2020-0601 CurveBall
CVE-2015-4000 Logjam
FREAK RSA_EXPORT downgrade
BREACH / CRIME compression oracles
Superfish / Komodia 2015 CA key leak
Dual_EC_DRBG backdoored RNG
Bellcore / Lenstra RSA-CRT fault
Bleichenbacher / Manger padding oracles
FAQ

Frequently asked questions

Is this real TLS, or a simplified simulation?

Real. Where a supported library (rustls) has no config knob for the historical misconfiguration being taught — finite-field DHE, RSA_EXPORT, explicit-curve ECDHE, static-key KEMTLS — the platform hand-rolls a genuine TLS 1.2/1.3 engine instead: real record layer, real key schedule, real Finished messages. Every algorithm in the lab passes its official Known-Answer Tests — ML-KEM-768 (FIPS 203), ML-DSA-65 (FIPS 204), HQC and SLH-DSA-SHAKE-128f (FIPS 205) against the NIST ACVP / KAT vectors, and the classical primitives (SHA-2/SHA-3, SHAKE, HMAC, HKDF, AES-GCM, PBKDF2, RSA PKCS#1 v1.5 & PSS, DSA, ECDSA P-256) against their FIPS / RFC test vectors. Keys and signatures therefore interoperate with any standard library.

Do I need to already know the underlying cryptography?

You should be comfortable with TLS fundamentals, a terminal, and modular arithmetic. Levels 1-2 explain enough as you go; Levels 3-4 assume you can follow a lattice-reduction or index-calculus argument.

What do I actually submit to solve a challenge?

A 32-byte secret, as 64 hex characters, that the TLS client sent the server inside the (broken) TLS session. It's checked against an independent validator process, not against a static answer file.

Is the platform's own source code included?

No — deliberately. The student archive ships pre-built Docker images and a launcher script only, so nothing on disk hands you the answer by accident. That's separate from the challenge solutions themselves, which are included in full.

Are solutions really included?

Yes, in the full lab. The manual explains the technique behind every one of the 125 challenges, and the solutions/ folder holds a working solution script for them. The free version includes the manual for its 12 challenges but no solution scripts. Try first — the solution is there the moment you want it.

Does this run on macOS or Windows?

Yes: it runs on Linux, macOS and Windows. There are two ways to run it: on the host (run_challenges.sh), which needs a rootful Linux Docker daemon whose bridge network the host can reach directly; or in the browser (run_in_browser.sh), which runs the whole lab — every attack tool and a full Wireshark desktop — inside one container you open at http://localhost:6080. In short, by platform:

  • Linux — run it directly, or use browser mode. The bundle ships native x86-64 (any Intel or AMD 64-bit CPU; Docker amd64) and Arm64 (aarch64 — Apple Silicon, Arm servers) images and auto-selects the set matching your CPU, so Arm Linux works out of the box with no emulation.
  • Windows — install WSL2 with a Linux distribution, install Docker Engine inside it, and run the lab from the WSL shell exactly as on Linux.
  • macOS — use browser mode: Docker runs in a VM, so the host can’t reach the lab network directly. Needs a Docker with BuildKit — Docker Desktop and OrbStack include it; with colima run brew install docker-buildx first.

Full step-by-step instructions for each platform are in the manual included with the download.

How do I install the tools the challenges and solutions need?

Browser mode needs nothing but Docker — every attack tool is already inside the container. For host mode, both the full lab and the free version ship two helper scripts at the root of the archive:

  • ./install_host_deps.sh installs everything in one command — the rootful Docker engine, the MITM and scanning tools (arpspoof, iptables, tcpdump, nmap, Wireshark) and the Python libraries the solution scripts use. It auto-detects your package manager, so it works on Debian / Ubuntu (apt), Fedora / RHEL (dnf), openSUSE (zypper) and Arch / Manjaro (pacman). Pass --core-only to skip the large optional extras (Metasploit, mitmproxy, hydra).
  • ./check_requirements.sh is read-only: it reports what's installed and what's missing — and the exact install command for your distribution — without installing or changing anything. Run it any time to confirm you're ready.

In the free version both scripts are trimmed to just that bundle's smaller tool set. Prefer not to touch your host at all? Use browser mode (run_in_browser.sh) instead.

The manual included with the download explains all of this in full — every tool the lab and the solution scripts use, what each one is for, and how to detect what's missing and install it on your platform.

What's the resource footprint when it's running?

Small either way. The three core containers (server, client, validator) are FROM scratch, UPX-packed Rust binaries: about 4.8 MB combined for the full 125-challenge lab and 4.6 MB for the free 12-challenge version (download: about 6 MB and 5 MB). RAM stays around 10 MB total either way — only one challenge runs at a time regardless of how many are compiled in, so RAM tracks the active challenge, not the build size. CPU usage is negligible outside brief bursts during key generation at startup. The launcher caps each container at 2 GB RAM / 2 CPU cores as a safety ceiling, not a requirement — real usage is a small fraction of that. One documented exception: server_rsa_hundreds_servers_tls12 spawns 501 parallel RSA-2048 key generations at startup and needs --cpus=8 to finish in reasonable time. The browser mode is heavier: its workstation image (desktop, Wireshark, attack tools, Docker) is built on your machine the first time you start it, needs an internet connection for that build, and takes about 3 GB of disk.

What if I get stuck?

Every challenge's write-up in the manual explains the vulnerability class and the technique that solves it, and the included solution script is there if you'd rather see it worked end to end. For anything beyond that, write to support@cryptoschola.com.

Ready to see it running?

Try 12 of the 125 challenges free, or get the full lab.