external/anthropic-cybersecurity-skills/skills/migrating-to-post-quantum-cryptography/SKILL.md
Inventory cryptography, deploy hybrid X25519 and ML-KEM, and prioritize harvest-now-decrypt-later data.
npx skillsauth add seikaikyo/dash-skills migrating-to-post-quantum-cryptographyInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Scope and Authorization: This skill describes defensive cryptographic-migration engineering on systems you own or operate. Cryptographic discovery scanning can touch sensitive key material and production traffic — run inventory tooling only with authorization and in line with your organization's change-management and data-handling policies.
A cryptographically relevant quantum computer (CRQC) running Shor's algorithm will break the public-key cryptography that secures almost all of today's communications and signatures: RSA, finite-field and elliptic-curve Diffie-Hellman (DH/ECDH), and ECDSA. Symmetric primitives (AES) and hashes (SHA-2/3) are only weakened (Grover gives a quadratic speedup, mitigated by larger key/output sizes), but asymmetric algorithms are catastrophically broken. The most urgent threat is harvest-now, decrypt-later (HNDL): adversaries capturing encrypted traffic today to decrypt once a CRQC exists, which puts long-lived secrets (health records, state secrets, intellectual property, root-of-trust keys) at risk now.
On 13 August 2024 NIST finalized the first post-quantum standards: FIPS 203 (ML-KEM, Module-Lattice KEM, formerly CRYSTALS-Kyber) for key establishment, FIPS 204 (ML-DSA, Module-Lattice digital signatures, formerly CRYSTALS-Dilithium), and FIPS 205 (SLH-DSA, the stateless hash-based signature scheme SPHINCS+). The migration playbook (NIST SP 1800-38, Migration to Post-Quantum Cryptography) is: (1) build a cryptographic inventory / CBOM, (2) prioritize by HNDL exposure and crypto-agility, (3) deploy hybrid schemes (a classical algorithm AND a PQC algorithm combined, e.g. X25519MLKEM768) so a break in either leg does not compromise the session, and (4) re-key and rotate.
This skill maps to ATT&CK T1573 – Encrypted Channel: the same cryptographic channels adversaries abuse for stealthy C2 are the channels defenders must make quantum-resistant; understanding the algorithms in use is foundational to both attack detection and defensive migration. The NIST CSF outcome is PR.DS-02 (data-in-transit protection) — and by extension data-at-rest for HNDL-sensitive stores.
X25519MLKEM768) on TLS endpoints, VPNs, or SSH.openssl version # expect 3.5.0+
openssl list -kem-algorithms | grep -i mlkem
openssl list -signature-algorithms | grep -i mldsa
git clone https://github.com/open-quantum-safe/liboqs && \
cmake -S liboqs -B liboqs/build && cmake --build liboqs/build && \
sudo cmake --install liboqs/build
git clone https://github.com/open-quantum-safe/oqs-provider && \
cmake -S oqs-provider -B oqs-provider/_build && \
cmake --build oqs-provider/_build && \
sudo cmake --install oqs-provider/_build
python3 -m pip install cryptography
cdxgen, or cbomkit-theia for container/directory crypto discovery.X25519MLKEM768 key exchange on a TLS endpoint.| ID | Official Technique Name | Relevance | |----|------------------------|-----------| | T1573 | Encrypted Channel | Migration secures the encrypted channels (TLS/VPN/SSH) that protect data in transit; cryptographic inventory of these channels also underpins detection of adversary-controlled encrypted C2. | | T1573.002 | Encrypted Channel: Asymmetric Cryptography | RSA/ECDH key exchange — the exact asymmetric primitives broken by a CRQC and replaced by ML-KEM hybrids. | | T1573.001 | Encrypted Channel: Symmetric Cryptography | AES and other symmetric ciphers; quantum-weakened by Grover, mitigated by 256-bit keys rather than replacement. |
openssl version
# List quantum-safe KEMs and signatures available in this OpenSSL build
openssl list -kem-algorithms | grep -Ei 'mlkem|kyber'
openssl list -signature-algorithms | grep -Ei 'mldsa|dilithium|slhdsa|sphincs'
openssl list -tls-groups 2>/dev/null | grep -Ei 'mlkem'
If using oqs-provider on OpenSSL 3.0–3.4, activate it in openssl.cnf:
[provider_sect]
default = default_sect
oqsprovider = oqsprovider_sect
[default_sect]
activate = 1
[oqsprovider_sect]
activate = 1
Generate a CycloneDX CBOM from a code repo or container with cbomkit-theia / cdxgen:
# Directory / container image crypto discovery
cbomkit-theia dir ./myapp --output cbom.json
# or with cdxgen (Java keystores, certs, source-level algorithms)
cdxgen -t java --include-crypto -o cbom.json ./myapp
Enumerate TLS algorithms and certificate signature schemes across live endpoints with the helper agent.py scan (below), and the public-key strength of any certificate:
openssl x509 -in server.crt -noout -text | grep -E 'Signature Algorithm|Public Key'
For each inventoried asset, record: algorithm, key size, where the key lives, data sensitivity, and data lifetime. Prioritize migration where data_lifetime_years + migration_time > years_until_CRQC (Mosca's inequality). Long-lived confidential data over public networks ranks highest; ephemeral internal traffic ranks lower. Hash-based signature roots-of-trust (firmware signing) are also high priority because they protect long-lived trust anchors.
# ML-KEM-768 (key establishment) keypair
openssl genpkey -algorithm ML-KEM-768 -out mlkem768.key
# OpenSSL 3.0-3.4 + oqs-provider uses lowercase 'mlkem768'
# openssl genpkey -algorithm mlkem768 -out mlkem768.key
# ML-DSA-65 (signature) keypair
openssl genpkey -algorithm ML-DSA-65 -out mldsa65.key
openssl pkey -in mldsa65.key -pubout -out mldsa65.pub
# Self-signed ML-DSA-65 certificate for testing
openssl req -new -x509 -key mldsa65.key -out mldsa65.crt -days 365 \
-subj "/CN=pqc-test.example.com"
openssl x509 -in mldsa65.crt -noout -text | grep -A1 'Signature Algorithm'
echo "firmware-image-v2.bin" > artifact.txt
openssl dgst -sign mldsa65.key -out artifact.sig artifact.txt
openssl dgst -verify mldsa65.pub -signature artifact.sig artifact.txt
# -> "Verified OK"
Run a TLS 1.3 server and force the hybrid group X25519MLKEM768 (classical X25519 + ML-KEM-768):
# Server (use a classical or ML-DSA cert/key)
openssl s_server -accept 4433 -www -tls1_3 \
-cert mldsa65.crt -key mldsa65.key -groups X25519MLKEM768
# Client — negotiate the hybrid group and confirm it was used
openssl s_client -connect localhost:4433 -tls1_3 -groups X25519MLKEM768 \
</dev/null 2>/dev/null | grep -E 'Negotiated|Server Temp Key|Cipher'
For external endpoints, confirm support against a public PQC test server:
openssl s_client -groups X25519MLKEM768 -tls1_3 -connect pq.cloudflareresearch.com:443 </dev/null
Configure the web server / load balancer to offer the hybrid group while keeping classical fallback for old clients. NGINX with OpenSSL 3.5+:
server {
listen 443 ssl;
ssl_protocols TLSv1.3;
ssl_ecdh_curve X25519MLKEM768:X25519:secp256r1; # hybrid first, classical fallback
ssl_certificate /etc/nginx/certs/server.crt;
ssl_certificate_key /etc/nginx/certs/server.key;
}
Reload and verify with the s_client command from step 7 against the live host.
Centralize algorithm selection (config, not code), record key/cert expiry, and schedule re-keying. Re-run the inventory (step 2) on a cadence to confirm no quantum-vulnerable-only algorithms remain on prioritized assets, and track residual RSA/ECDH usage to zero on high-HNDL paths.
| Tool / Resource | Purpose | Link | |-----------------|---------|------| | FIPS 203 (ML-KEM) | KEM standard | https://csrc.nist.gov/pubs/fips/203/final | | FIPS 204 (ML-DSA) | Signature standard | https://csrc.nist.gov/pubs/fips/204/final | | FIPS 205 (SLH-DSA) | Hash-based signature standard | https://csrc.nist.gov/pubs/fips/205/final | | NIST SP 1800-38 | Migration practice guide / crypto discovery | https://www.nccoe.nist.gov/crypto-agility-considerations-migrating-post-quantum-cryptographic-algorithms | | OpenSSL 3.5 | Native ML-KEM/ML-DSA/SLH-DSA + hybrid groups | https://www.openssl.org | | oqs-provider / liboqs | PQC for OpenSSL 3.0–3.4 | https://github.com/open-quantum-safe/oqs-provider | | CycloneDX CBOM | Cryptography Bill of Materials spec | https://cyclonedx.org/capabilities/cbom/ | | CBOMkit / cbomkit-theia | Crypto discovery & CBOM generation | https://github.com/cbomkit/cbomkit-theia |
| Classical (broken/weakened) | Quantum-safe replacement | Standard | Use | |-----------------------------|--------------------------|----------|-----| | RSA / ECDH / DH key exchange | ML-KEM-512/768/1024 (hybrid: X25519MLKEM768) | FIPS 203 | Key establishment | | RSA / ECDSA / EdDSA signatures | ML-DSA-44/65/87 | FIPS 204 | General signatures | | (backup signature) | SLH-DSA (SPHINCS+) | FIPS 205 | Conservative/firmware signing | | AES-128 | AES-256 | FIPS 197 | Symmetric (Grover-hardened) | | SHA-256 | SHA-384/512, SHA-3 | FIPS 180-4/202 | Hashing |
X25519MLKEM768 key exchange negotiated and confirmed on a test endpoint.development
拋棄式 HTML mockup 比稿:產出 2 到 3 個設計立場不同的變體(密度 / 版式 / 強調軸,不是換色),各附取捨說明,最後給有立場的對比結論。適用:「畫個草圖」「比較 A 版 B 版」「先看方向再做」「給我看幾種做法」。要 production 元件或設計已定案時不適用。
tools
需求不明時的意圖萃取訪談:一次一題、每題附上自己的猜測、聽出「真正想要 vs 覺得應該要」,直到能預測使用者反應(約 95% 信心)才動工。適用:需求缺少對象 / 動機 / 成功標準 / 約束,或使用者點名「訪談我」「先確認一下」「我們確定嗎」。明確自足的指示、純資訊查詢、機械性操作不適用。
development
對非平凡決策啟動新鮮 context 對抗審查(找碴不背書),在修正還便宜的時候抓出錯誤方向。適用:高風險改動(production、資安敏感邏輯、不可逆操作)、不熟的程式碼、要宣稱「這樣是安全的 / 可行的」之前。機械性操作與一行修改不適用。
testing
Reference for writing and editing agent skills well — the vocabulary and principles that make a skill predictable. Consult when authoring, reviewing, or pruning a SKILL.md.