Files
WRNexusJS/docs/SECURITY-SUPPORT-MATRIX.md
Clintchiz 586a6db8ff
Quality / quality (ubuntu-latest) (push) Failing after 21s
Quality / quality (windows-latest) (push) Canceled after 0s
release: WRNexusJS 0.8.0
2026-08-02 23:18:51 +05:30

3.9 KiB

Security support and threat model matrix

WRNexus security controls are defense layers, not substitutes for TLS, secret management, database isolation, or operational monitoring. This matrix defines the supported 0.8 contracts and the boundary application owners retain.

Area Framework guarantee Application/operator responsibility
Sessions and passwords Rotating sessions, idle/absolute expiry, revocation, password hashing policy, trusted-device controls TLS, durable session storage, breach response, account recovery policy
MFA and passkeys One-use OTP/recovery state, TOTP counter replay defense, WebAuthn challenge consumption, RP/origin provider inputs, signature-counter regression checks Use a conformant WebAuthn provider, configure the exact public origin/RP ID, transactional credential storage
OAuth/OIDC State consumption, PKCE, safe return URLs, strict issuer discovery, HTTPS endpoints, RS256 JWKS rotation, nonce/audience/azp/at_hash validation Register exact redirect URIs, protect client secrets, select trusted issuers and scopes
JWT HS256 local keyrings and bounded HTTPS RS256 JWKS caches with kid rollover Rotate/retire keys deliberately, set issuer/audience/expiry/max-age, revoke refresh families
CAPTCHA and rate limits Expiring one-use challenges, action/session binding, bounded create/verify endpoints Treat CAPTCHA as one signal and keep route/account/network rate limits
Encrypted HTTP AEAD, request context binding, age checks, optional one-use replay stores Keep HTTPS, use a shared replay store across instances, rotate server-managed keys
Authorization Explicit policies/roles and deny responses Model tenant/object ownership and test privilege boundaries
Browser/runtime Production CSP, HSTS, MIME-sniffing, referrer/permissions policies and safe hydration serialization Configure allowed origins/sources, terminate TLS correctly, review third-party scripts

AUTH_SECURITY_EVENT_TYPES is the stable machine-readable authentication audit event vocabulary. Events include an ID, timestamp, severity and available actor, user, session, IP, user-agent and structured data. Consumers should alert on critical events and retain warning/critical records according to their policy; never place credentials or raw tokens in event data.

Protocol and adversarial tests cover replay, expired state, counter regression, key rotation, algorithm/key-type confusion, issuer/audience/nonce mismatch, unsafe return URLs, request binding, rate limiting, and built production response headers. Live provider certification remains an integration responsibility because provider metadata and credential policies are deployment-specific.