Methodology v3.1.0

The framework scores the chains in the LayerQu evaluation set on their readiness to keep working after a cryptographically relevant quantum computer arrives. The headline output is Migration Stage on a 0 to 5 scale. The secondary output is the Quantum Readiness Index (QRI), 0 to 100. Each chain is also flagged against five gates and six caps that surface specific failure modes a non-cryptographer can read at a glance.

v3.1.0 is the third revision of the framework, ratified 2026-05-01. It supersedes v3.0 (ratified 2026-04-24) and v2 (2026-04-17). The change log is at the bottom of this page.

Two structural changes matter most. First, Migration Stage replaced QRI as the headline because architectural elegance is not deployment, and a chain can score well on paper while shipping nothing observable on mainnet. Second, Dimension 2 was split into Forge (signature forgeability after Shor) and Decrypt (recovery of harvested ciphertext after Shor) because the two threats have different timelines and different mitigations. Dimension 3 was split into Anonymity and Confidentiality subtotals for privacy-focused chains because a chain can hide the transaction graph while leaving every encrypted note exposed under Shor. Two new gates were added around hybrid composition, which is the only deployment pattern that survives a CRQC during transition.

Six principles

• Public artifacts only. GitHub commits, governance proposals, audit reports, conference talks, peer-reviewed papers, foundation blogs. No private interviews on the scoring path.
• Primitives named, not categories. Not "signatures." ECDSA over secp256k1. Not "ZK." Halo2-KZG over BN254 (PSE / pairing-based) versus Halo2-IPA over Pasta (Zcash-style, non-pairing); the distinction matters for Shor exposure.
• Hybrid is mandatory. Stage 5 is hybrid-PQ, not pure-PQ. A pure-replacement migration that depends on a single PQ family is a single point of failure, and the standardisation track has revisable cryptanalysis. Hybrid composition is the design that survives classical and quantum adversaries during transition.
• Stage over QRI. A chain can be architecturally elegant and operationally absent. Migration Stage measures what has shipped; QRI measures what could ship.
• Washing discount. Chains that talk more than they ship get penalized, not rewarded. The ratio of announcements to shipped code is published explicitly.
• Independence. No paid listings, no sponsored upgrades, no chain foundations on the reviewer list. The framework is the framework regardless of who is reading.

Migration Stage 0 to 5, the headline output

The single number a non-cryptographer should read on a chain page. Stage answers: what has actually shipped?

Stage	Label	Test
0	Unaware	No public foundation statement on PQ posture; no plan; no spec.
1	Acknowledged	Public statement of intent; no architecture; no code.
2	Architected	Spec ratified or PQ code merged into client; mainnet feature-flagged off acceptable.
3	Piloted	Real PQ primitive signing real protocol data on mainnet, opt-in or parallel; not majority.
4	Hybrid shipped	Hybrid composition (Gate 1a-Sig PASS) on mainnet with non-trivial traffic.
5	Hybrid majority	Majority of mainnet traffic under hybrid PQ; classical sunset declared and dated.

Caps reduce maximum Stage. Milestone-Discipline Cap pulls Stage to 2 if 5d (published dated milestones) is voided. Supply-Chain Cap pulls Stage to 3 if 3+ of 4 vendor tiles lack PQC roadmap.

QRI 0 to 100, the granular index

The Quantum Readiness Index is the weighted roll-up across the seven dimensions. It is published as a secondary stat next to Stage. The QRI captures gradations, a Stage-2 chain at QRI 25 looks meaningfully different from a Stage-2 chain at QRI 41, but it is not the primary call.

QRI is bounded by the six caps (see below). A chain with strong architecture (Dim 4) and weak deployment (Dim 5) will have a high raw QRI capped at 70 by the Architecture-Execution Gap Cap.

Bands, qualitative summary

Bands compress QRI into a label readers can quote without misrepresenting.

Band	QRI	Label
1	0–10	Unaware
2	11–20	Acknowledged
3	21–30	Planning
4	31–40	Architected
5	41–50	Prototyped
6	51–70	Transitioning
7	71–100	Deployed

The seven dimensions

Each dimension scores 0 to 100 internally. Weights vary by profile (see profiles).

#	Dimension	Sub-scores	What it measures
1	Cryptographic Exposure	1a–1e	Inventory of named primitives, Shor/Grover classification, family diversity, NIST category mapping, implementation quality.
2	Quantum Recovery Exposure	2a–2d (or 2a–2e for privacy)	Forge subtotal (active-key, cold-key, signature long-term validity) + Decrypt subtotal (HNDL, note-ciphertext payload).
3	Metadata, Anonymity & Confidentiality	3a–3e (or 3a–3f for privacy)	Anonymity subtotal (tx-graph, RPC concentration, bridge correlation, mixnet) + Confidentiality subtotal (retroactive deanon, content-payload encryption shelf life).
4	Migration Architecture	4a–4f	Crypto-agility, AA / key rotation, hard-fork track record, hybrid deployment readiness, stateful-hash management, BFT aggregation-path declaration.
5	Deployment Execution	5a–5f	Mainnet PQC traffic %, code in consensus client, validator adoption, dated milestones, washing delta, signature-footprint multiplier.
6	Supply Chain Vendor Readiness	6a–6d	Top-3 wallet, bridge, custodian, and RPC/HSM/TEE PQC roadmaps. Cross-tile weak-link cap applies.
7	Governance & Coordination	7a–7e	Validator/stake distribution, upgrade cadence under pressure, named coordination lead, adversarial-coordination precedent, canary/tripwire mechanism.

Forge / Decrypt + Anonymity / Confidentiality splits

v3.1.0 publishes two subtotals on Dim 2 because Forge (signature forgeability post-Shor) and Decrypt (HNDL ciphertext recovery) are different threats with different timelines. A chain can have low Decrypt exposure (no on-chain encrypted state) but catastrophic Forge exposure (every active address reveals its public key on first spend).

v3.1.0 publishes two subtotals on Dim 3 for privacy-focused chains, Anonymity (graph hiding) and Confidentiality (content hiding). The split surfaces the dominant privacy-chain finding: shielded-by-default chains can have strong Anonymity (50+ / 80) and zero Confidentiality (0/40) when every shielded primitive depends on Shor-vulnerable curves. Single-Q-day events decrypt the entire historical pool.

Three profiles

Different chain types absorb post-quantum risk differently. Three weight profiles.

Dimension	L1	rollup-L2	privacy-focused-chain
1 Cryptographic Exposure	15%	12%	12%
2 Quantum Recovery Exposure	10%	8%	10%
3 Metadata / Anonymity / Confidentiality	13%	8%	25% (split)
4 Migration Architecture	10%	15%	12%
5 Deployment Execution	22%	22%	18%
6 Supply Chain Vendor Readiness	22%	25%	18%
7 Governance & Coordination	8%	10%	5%
Privacy-focused-chain weights split Dim 3 into 15% Anonymity + 10% Confidentiality.

Five gates

Gates flag, they do not rank. A chain can fail any gate and still score well on the granular dimensions. The point is to surface conditions a non-cryptographer can read at a glance.

Gate	PASS condition	Failure signal
Gate 1a, Hybrid signature	Hybrid composition (AND or OR-with-commit-to-hash) documented at signature path	No composition documented; pure-replacement deployment counts as FAIL
Gate 1a, Hybrid KEM	Hybrid PQ KEM at validator transport (concatenated shared-secret per IETF hybrid TLS draft)	Classical KEM only at consensus / RPC / peer transport
Gate 1b, Commit-to-hash	For OR-composition only: commit-to-hash-of-both-pubkeys documented	OR-composition without commit; N/A if no OR-composition
Gate 2, Evidence reconstruction	Every sub-score is backed by ≥ 3 source artifacts (foundation specifications, code commits, third-party analyses) held in the internal audit repository, granted on request via /audit-access	Any sub-score has < 3 artifacts; failure triggers Evidence-Density Cap
Gate 3, Primitive naming	Every sub-score names specific primitives with mechanism (curve, hash, KDF, AEAD)	Category-level labels ("signatures", "ZK") at any sub-score

Six caps

Caps are floors on the headline outputs. Granular dimension scores are unaffected; the QRI and Migration Stage are bounded.

Cap	Trigger	Effect
Mainnet-Traffic Cap	5a (mainnet PQC traffic %) < 20%	QRI ≤ 60
Architecture-Execution Gap Cap	Dim 4 − Dim 5 ≥ 25 points	QRI ≤ 70
Milestone-Discipline Cap	5d (published dated milestones) voided to 0; voided when 5a = 0	Migration Stage ≤ 2
Evidence-Density Cap	Any dim has < 3 evidence artifacts at sub-score level	That dim's weighted contribution discounted 50%
Supply-Chain Cap	3+ of 4 vendor tiles (wallet/bridge/custodian/RPC) lack top-3 PQC roadmap	Migration Stage ≤ 3
Cryptographic-Diversity Cap	Lattice-monoculture in PQ deployment or roadmap (no hash-based or code-based fallback at signature path)	QRI ≤ 60

Caps lower the headline outputs; they do not replace the granular sub-scores. Lowest binding cap wins.

X + Y vs Z, the timing test

Demoted from headline to secondary signal in v3.1.0, kept because it is the right question to ask: can a chain finish migrating before the threat lands?

• X: signature shelf life, how many years a signature must remain unforgeable. For privacy-focused chains, also X_priv: note-encryption shelf life.
• Y: years to Stage 5 from current state, accounting for governance cadence, vendor readiness, and gate failures.
• Z10: 10% probability year for a cryptographically-relevant quantum computer (≈ 2030 per Global Risk Institute 2025).
• Z25: 25% probability year (≈ 2035).
• Z compliance: NIST IR 8547 (2030 deprecation, 2035 disallowance), CNSA 2.0, BSI Migrationsplan, EU NIS2 / DORA.

If X + Y > Z10 the chain is in Crisis Zone, it cannot finish migrating before the lower-probability threat. If X + Y > Z25 the chain is Outside risk window. If X + Y > Z compliance the chain is Outside compliance window.

Announcement-to-shipped ratio

A ratio of PQC-related claims in trailing-12-month public communications to PQC-related shipped code on mainnet under named primitive.

Ratio	Tag	Effect
< 1.5	none	No deduction
1.5 – 2.0	>1.5 deduction	10-point deduction at 5e
2.0 – 5.0	QRI cap 65	QRI ≤ 65 (does not bind unless raw QRI > 65)
> 5.0 (or undefined when shipped = 0)	narrative-only	10-point deduction + QRI cap 65 + tag rendered on chain card

Burn-vs-rescue policy

v3.1.0 requires every chain card to declare a burn-vs-rescue policy: what happens to dormant balances on the unmigrated classical scheme once the PQ scheme is dominant?

Option	Description	Example
(a) Freeze / burn	Sunset date after which classical signatures are no longer accepted; stuck balances are unspendable	Bitcoin BIP-361 (Draft), Phase B at +5y invalidates ECDSA/Schnorr
(b) STARK rescue	Zero-knowledge proof of pubkey ownership rescues balances after sunset	BIP-361 Phase C zk-recovery (research)
(c) Hybrid client-layer	Wallet-level dual-signing during transition	n/a
(d) Hourglass / rate-limit canary	Cap on per-block spend from quantum-vulnerable addresses	Bitcoin Hourglass V2 (proposal)
(e) Optional migration	No forced sunset; users opt in voluntarily; long-tail balances remain at risk indefinitely	Adam Back position; majority of L1s by default
(f) Undeclared	No policy on file	Most chains in the v3.1.0 set

Declaring (f) "undeclared" is itself a signal. The absence of a policy is a Dim 7 governance finding.

Source-disagreement disclosure

Every chain card publishes a source-disagreement disclosure, material divergences among authoritative sources (foundation vs independent press, methodology vs alternative weighting, primary spec vs third-party interpretation). Plus a delta-QRI under the alternative weighting.

This is a transparency requirement, not a hedge. The disclosure surfaces where a reasonable evaluator could land at a different number under a defensible alternative reading. The methodology call is what the card carries; the alternative is what the next reviewer should test.

The audit packet

Every published score is built from a packet of foundation specifications, source commits, and third-party analyses. The packet is held in our internal audit repository. Public chain cards on layerqu.com display the substantive claim and the source category, not the evidence URL list, because a single broken or stale link in a public scorecard does more damage to the reader than the link adds in transparency.

Reviewers who need the underlying packet for verification request access at /audit-access. Access is granted on a case-by-case basis to chain foundations, regulators, standards bodies, researchers, and institutional reviewers with a stated verification need.

If a reviewer flags a score they cannot reconstruct from the packet, we retract the score and keep the flag visible until the issue is resolved.

Common questions

---

LayerQu v3.1.0 methodology, ratified 2026-05-01, supersedes v3.0 (2026-04-24), v2 (2026-04-17), and ChainScreen v1. Next review 2026-08-01.