AgentRules.org
RulesStacksClientsAuditorPrunerSelf-HealRFCsBenchmarkCLICI/CDAuthorsSubmit
This specification is maintained by the AgentRules Working Group. Have enhancements or want to claim maintainership?Submit Specification →
securitycritical PriorityID: webauthn-l3-prf-deterministic-key-derivation-guardrail

WebAuthn Level 3 PRF Extension Deterministic Key Derivation Invariant

Mandates client-side hardware-bound deterministic key derivation via the WebAuthn Level 3 PRF (Pseudo-Random Function) extension. Enforces strict zero-knowledge cryptographic boundaries ensuring raw PRF outputs and derived symmetric encryption keys never traverse network boundaries or persist into non-volatile client storage.

Prohibited Anti-Pattern: Exfiltrating PRF outputs to server endpoints or persisting raw derived symmetric keys in client storage.

Export Ready-to-Drop Configuration

Select format below to copy code or download file directly.

---
description: WebAuthn Level 3 PRF Extension Deterministic Key Derivation Invariant - Mandates client-side hardware-bound deterministic key derivation via the WebAuthn Level 3 PRF (Pseudo-Random Function) extension. Enforces strict zero-knowledge cryptographic boundaries ensuring raw PRF outputs and derived symmetric encryption keys never traverse network boundaries or persist into non-volatile client storage.
globs: src/**/*.{ts,tsx,js,mjs}
alwaysApply: true
verified_by: "agentrules.org"
license: "CC-BY-4.0"
source: "https://agentrules.org/rules/webauthn-l3-prf-deterministic-key-derivation-guardrail"
---

# WebAuthn Level 3 PRF Extension Deterministic Key Derivation Invariant

> **Severity**: CRITICAL | **Category**: security
> **Anti-Pattern Warning**: Exfiltrating PRF outputs to server endpoints or persisting raw derived symmetric keys in client storage.

## System Constraints
You are a Principal Security Architect enforcing the WebAuthn Level 3 PRF (Pseudo-Random Function) Extension Deterministic Key Derivation Invariant. You must reject any code where WebAuthn PRF outputs, raw PRF seed values, or symmetric encryption keys derived from hardware authenticators are transmitted to a remote server, serialized in HTTP request payloads, logged to telemetry, or stored in non-volatile browser storage (e.g., localStorage, sessionStorage, unencrypted IndexedDB). Ensure that PRF evaluations occur exclusively on the client, inputs are cryptographically isolated with domain-separated salts, and derived keys remain as non-extractable CryptoKey handles within the Web Crypto API.

## Strict Guidelines
- Engineers MUST ALWAYS isolate raw PRF outputs (AuthenticationExtensionsPRFOutputs) strictly to ephemeral client-side memory and immediately ingest them into SubtleCrypto as non-extractable CryptoKey handles.
- Engineers MUST NOT include PRF evaluation outputs or derived symmetric keys in credential assertion payloads transmitted back to relying party (RP) server endpoints.
- Code MUST sanitize client extension results using an explicit allowlist before sending WebAuthn assertion verification payloads to backend endpoints, stripping prf output keys.
- Services MUST provide application-specific, domain-separated cryptographic salts (minimum 32 bytes) for PRF evaluation inputs, ensuring identical authenticators produce distinct key spaces across operational contexts.
- Components NEVER persist raw PRF outputs or unencrypted symmetric keys in Web Storage APIs (localStorage, sessionStorage) or unencrypted client databases.
- Key derivation pipelines MUST utilize HKDF (RFC 5869) with SHA-256 or SHA-512 to expand the raw authenticator PRF secret into symmetric application keys (e.g., AES-GCM-256) marked with extractable: false.

## Bad vs Good Implementations
### Hardware-Bound PRF Symmetric Key Derivation with Server-Side Sanitization

#### Bad Antipattern (AI Common Hallucination)
> **Why it fails**: The implementation uses an insecure non-domain-separated salt and directly serializes the entire WebAuthn assertion payload—including the raw PRF output extension results—over the network to the server, violating the zero-knowledge hardware-isolation invariant.
```typescript
import { startAuthentication } from '@simplewebauthn/browser';

export async function unlockUserVault(options: any) {
  // ANTI-PATTERN: Requesting PRF evaluation and blindly returning full assertion to server
  const assertion = await startAuthentication({
    ...options,
    extensions: {
      prf: {
        eval: { first: new Uint8Array([1, 2, 3, 4]) } // Insecure static salt
      }
    }
  });

  // CRITICAL SECURITY FLAW: Sending PRF output containing hardware key derivation secrets to server
  const response = await fetch('/api/vault/verify-and-unlock', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify(assertion) // Leaks assertion.clientExtensionResults.prf to network boundary
  });

  return response.json();
}
```

#### Production Standard (Correct)
> **Standard**: Applies a minimum 32-byte domain-separated salt, parses the PRF extension output exclusively in client memory, immediately zeroes out raw buffers, imports into WebCrypto as a non-extractable HKDF root key, derives a non-extractable AES-GCM key, and strips the PRF result before transmitting the authentication assertion to the server.
```typescript
import { startAuthentication } from '@simplewebauthn/browser';
import type { AuthenticationResponseJSON } from '@simplewebauthn/types';

interface AuthenticatorAssertionWithPRF extends AuthenticationResponseJSON {
  clientExtensionResults: AuthenticationResponseJSON['clientExtensionResults'] & {
    prf?: { results?: { first: string } };
  };
}

export async function deriveHardwareBoundVaultKey(
  options: any,
  domainSalt: Uint8Array
): Promise<CryptoKey> {
  if (domainSalt.byteLength < 32) {
    throw new Error('Security Invariant Violation: PRF salt must be >= 32 bytes');
  }

  const assertion = (await startAuthentication({
    ...options,
    extensions: {
      ...options.extensions,
      prf: { eval: { first: domainSalt } }
    }
  })) as AuthenticatorAssertionWithPRF;

  const rawPrfBase64 = assertion.clientExtensionResults?.prf?.results?.first;
  if (!rawPrfBase64) {
    throw new Error('Authenticator did not return PRF extension output');
  }

  // Convert URL-safe base64 / binary string to Uint8Array buffer
  const rawPrfBytes = Uint8Array.from(atob(rawPrfBase64.replace(/-/g, '+').replace(/_/g, '/')), c => c.charCodeAt(0));

  // Derive ephemeral root key material in WebCrypto
  const prfIkmKey = await window.crypto.subtle.importKey(
    'raw',
    rawPrfBytes,
    { name: 'HKDF' },
    false, // Non-extractable
    ['deriveKey']
  );

  // Zero out raw byte buffer immediately from heap
  rawPrfBytes.fill(0);

  // Derive target AES-GCM symmetric key with strict zero-knowledge parameters
  const derivedSymmetricKey = await window.crypto.subtle.deriveKey(
    {
      name: 'HKDF',
      hash: 'SHA-256',
      salt: domainSalt,
      info: new TextEncoder().encode('app-client-vault-v1')
    },
    prfIkmKey,
    { name: 'AES-GCM', length: 256 },
    false, // Must NEVER be extractable
    ['encrypt', 'decrypt']
  );

  // Sanitize verification payload before transmitting to RP server
  const sanitizedAssertion: AuthenticationResponseJSON = {
    ...assertion,
    clientExtensionResults: {
      ...assertion.clientExtensionResults,
      prf: undefined // Strips raw PRF outputs from network boundary
    }
  };

  const verificationResponse = await fetch('/api/webauthn/assertion/verify', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify(sanitizedAssertion)
  });

  if (!verificationResponse.ok) {
    throw new Error('Relying party assertion verification failed');
  }

  return derivedSymmetricKey;
}
```
Recommended MCP Servers (mcp-serverlist.com)

Equip agent with system tools to verify code for this stack.

Security AST Scanner MCP
Scans client-side WebAuthn implementations for cryptographic leaks, missing WebCrypto parameter constraints, and telemetry exposure.
Recommended Agent Skills (askills.dev)

Ready-to-use executable skill functions tailored to this stack.

WebAuthn PRF Security Linter
Static code analysis tool checking for non-extractable CryptoKey constraints and network serialization of WebAuthn extension results.
Architectural Specification & Invariant Rationale

Architectural Invariant Design: WebAuthn Level 3 PRF Extension Deterministic Key Derivation Invariant

WebAuthn Level 3 PRF Extension Deterministic Key Derivation Invariant defines an industrial-grade behavioral standard for security engineering. Mandates client-side hardware-bound deterministic key derivation via the WebAuthn Level 3 PRF (Pseudo-Random Function) extension. Enforces strict zero-knowledge cryptographic boundaries ensuring raw PRF outputs and derived symmetric encryption keys never traverse network boundaries or persist into non-volatile client storage. Modern software architectures require strict adherence to deterministic patterns to guarantee high availability, maintainability, and zero security vulnerabilities.

Framework Paradigm Shift & Drift MechanicsLLMs are trained on millions of open-source repositories containing outdated code, legacy idioms, and unpatched antipatterns. When coding assistants generate implementations for this stack, they default to common statistical tokens: Exfiltrating PRF outputs to server endpoints or persisting raw derived symmetric keys in client storage. Without an explicit system instruction file, developers spend hours debugging silent failures.
Failure Modes & Production Blast Radius

Hazard Analysis: What Breaks Without This Rule

The following failure modes have been cataloged from hundreds of hours of production code reviews where AI assistants produced non-compliant implementations:

Failure ModeAI Symptom (The Anti-Pattern)SeverityProduction Impact & Remedy
AI Hallucination & Deprecation DriftExfiltrating PRF outputs to server endpoints or persisting raw derived symmetric keys in client storage.Critical
Violating this industrial standard introduces regression bugs, security risks, and unexpected runtime failures in production.
Enforce explicit behavioral system directives across all AI coding assistants in your repository.
Silent Architecture InconsistencyAI assistants produce mixed paradigms across files, creating fragmented codebase conventions.Medium
Increases cognitive load during code reviews and leads to maintenance debt across engineering teams.
Pin project conventions in .cursor/rules, CLAUDE.md, and AGENTS.md so all team agents follow identical guardrails.
Architectural Directives in Detail

Step-by-Step Production Directives

Each rule constraint below is injected into the AI agent's system prompt to establish deterministic boundaries:

1

Engineers MUST ALWAYS isolate raw PRF outputs (AuthenticationExtensionsPRFOutputs) strictly to ephemeral client-side memory and immediately ingest them into SubtleCrypto as non-extractable CryptoKey handles.

Engineers MUST ALWAYS isolate raw PRF outputs (AuthenticationExtensionsPRFOutputs) strictly to ephemeral client-side memory and immediately ingest them into SubtleCrypto as non-extractable CryptoKey handles.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
2

Engineers MUST NOT include PRF evaluation outputs or derived symmetric keys in credential assertion payloads transmitted back to relying party (RP) server endpoints.

Engineers MUST NOT include PRF evaluation outputs or derived symmetric keys in credential assertion payloads transmitted back to relying party (RP) server endpoints.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
3

Code MUST sanitize client extension results using an explicit allowlist before sending WebAuthn assertion verification payloads to backend endpoints, stripping prf output keys.

Code MUST sanitize client extension results using an explicit allowlist before sending WebAuthn assertion verification payloads to backend endpoints, stripping prf output keys.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
4

Services MUST provide application-specific, domain-separated cryptographic salts (minimum 32 bytes) for PRF evaluation inputs, ensuring identical authenticators produce distinct key spaces across operational contexts.

Services MUST provide application-specific, domain-separated cryptographic salts (minimum 32 bytes) for PRF evaluation inputs, ensuring identical authenticators produce distinct key spaces across operational contexts.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
5

Components NEVER persist raw PRF outputs or unencrypted symmetric keys in Web Storage APIs (localStorage, sessionStorage) or unencrypted client databases.

Components NEVER persist raw PRF outputs or unencrypted symmetric keys in Web Storage APIs (localStorage, sessionStorage) or unencrypted client databases.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
6

Key derivation pipelines MUST utilize HKDF (RFC 5869) with SHA-256 or SHA-512 to expand the raw authenticator PRF secret into symmetric application keys (e.g., AES-GCM-256) marked with extractable

Key derivation pipelines MUST utilize HKDF (RFC 5869) with SHA-256 or SHA-512 to expand the raw authenticator PRF secret into symmetric application keys (e.g., AES-GCM-256) marked with extractable: false.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
Technical Questions & Answers

Frequently Asked Technical Questions

Common questions developers face when configuring AI coding assistants for this stack:

Q:Why is WebAuthn Level 3 PRF Extension Deterministic Key Derivation Invariant critical for AI coding assistants?

Mandates client-side hardware-bound deterministic key derivation via the WebAuthn Level 3 PRF (Pseudo-Random Function) extension. Enforces strict zero-knowledge cryptographic boundaries ensuring raw PRF outputs and derived symmetric encryption keys never traverse network boundaries or persist into non-volatile client storage. Without explicit behavioral boundaries, AI coding assistants (Cursor, Claude Code, Windsurf, Codex) rely on training data averages, which frequently produce deprecated, inefficient, or vulnerable code patterns.

Q:How does adopting this rule prevent AI hallucinations?

This rule explicitly declares what the AI must avoid (Exfiltrating PRF outputs to server endpoints or persisting raw derived symmetric keys in client storage.) and prescribes verified positive examples. Grounding the LLM with negative and positive pairs eliminates guesswork.

Q:Which AI coding environments are supported?

This specification is multi-platform compatible: Cursor (.cursor/rules/*.mdc & .cursorrules), Anthropic Claude Code (CLAUDE.md), Windsurf (.windsurfrules), Codex / Cline (AGENTS.md), and GitHub Copilot (.github/copilot-instructions.md).

Q:Can I combine this rule with other technology stacks?

Yes. Use the Interactive Stack Builder or download an Agent Pack to merge this rule with other frontend, backend, database, security, and DevOps rules without syntax collision.

Open Specification · CC BY 4.0·Free for Personal, Team & Commercial In-Repo Use

Citation & Open Source Attribution Guidelines

This specification is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0). You are completely free to adapt and share this rule in your workspace. If citing, including in public AI benchmark datasets, or redistributing, attribution to AgentRules.org is required.

Related SECURITY Rules

View all rules →
security
105

OpenFGA Graph Invariants: Cyclic Relation Prevention & Contextual Tuple Hardening

Enforces strict directed acyclic graph (DAG) topology validation during relationship mutation writes and mandates strict server-side schema verification for ephemeral contextual tuples in OpenFGA evaluations. Prohibits persistent storage of ephemeral request attributes and guards against recursive relation loops causing graph expansion denial-of-service.

security
81

WebAuthn Conditional UI Mediation, Nonce Atomicity, and Assertion Cryptographic Verification

Enforces zero-trust cryptographic validation for FIDO2/WebAuthn Passkey authentication with Conditional UI autofill. Requires atomic single-use challenge consumption to eliminate replay attacks, strict RP ID hash and origin binding, authenticatorData flag verification (User Presence and User Verification), and monotonic signature counter enforcement to detect authenticator cloning.

security
111

Enforce Sigstore Keyless OIDC Attestation and Admission Controller Verification

Enforces immutable supply-chain security using Sigstore Cosign keyless signing via ephemeral OIDC tokens issued by Fulcio and verified against the Rekor transparency log. Container workloads must be signed and attested at build time by digest, and admission controllers (such as Kyverno or Sigstore Policy Controller) must strictly validate the OIDC issuer and subject identity before allowing image execution.

1 Rule in Stack

WebAuthn