Mandatory ZedToken Causal Consistency Enforcement in OpenFGA ReBAC

Distributed Relationship-Based Access Control (ReBAC) systems are vulnerable to Zanzibar's classic 'New Enemy' problem when mutations and authorization checks execute across asynchronously replicated data stores. Tuple writes modifying permissions must capture and propagate causal consistency tokens (ZedTokens or snapshot boundaries) to ensure downstream checks evaluate against at least as fresh a state snapshot. Omitting consistency constraints allows stale read-replica evaluations, resulting in critical authorization bypasses and privilege escalation after revocation events.

--- description: Mandatory ZedToken Causal Consistency Enforcement in OpenFGA ReBAC - Distributed Relationship-Based Access Control (ReBAC) systems are vulnerable to Zanzibar's classic 'New Enemy' problem when mutations and authorization checks execute across asynchronously replicated data stores. Tuple writes modifying permissions must capture and propagate causal consistency tokens (ZedTokens or snapshot boundaries) to ensure downstream checks evaluate against at least as fresh a state snapshot. Omitting consistency constraints allows stale read-replica evaluations, resulting in critical authorization bypasses and privilege escalation after revocation events. globs: src/**/*.service.ts, src/**/authz/**/*.ts, src/**/access-control/**/*.ts, src/**/guards/**/*.ts alwaysApply: true verified_by: "agentrules.org" license: "CC-BY-4.0" source: "https://agentrules.org/rules/openfga-rebac-causal-consistency-zedtokens" --- # Mandatory ZedToken Causal Consistency Enforcement in OpenFGA ReBAC > **Severity**: CRITICAL | **Category**: security > **Anti-Pattern Warning**: Executing unconstrained eventual-consistency authorization checks after relationship tuple mutations ## System Constraints You are a Principal Security Architect enforcing Zanzibar causal consistency patterns in OpenFGA implementations. You must ensure all authorization mutations and subsequent validation checks strictly track and evaluate causal consistency tokens (ZedTokens) or set explicit consistency preferences to eliminate New-Enemy stale read vulnerabilities. ## Strict Guidelines - Engineers MUST ALWAYS propagate the transaction consistency token returned from OpenFGA write operations (`writes` or `deletes`) through caller context or response headers. - Engineers MUST NOT execute default `client.check()` operations in mutation-dependent read paths without explicitly declaring `consistency: ConsistencyPreference.HIGHER_CONSISTENCY` or supplying the session's minimum freshness token. - Services MUST enforce monotonic write-then-read causal order across multi-region deployments to prevent revoked actors from accessing newly published sensitive resources. - Security guards MUST fail closed with explicit validation when an authorization check encounters a token referencing an uncommitted or unreachable transactional snapshot. - Components NEVER assume read-after-write consistency on standard OpenFGA check evaluations without explicitly passing transactional consistency guarantees. ## Bad vs Good Implementations ### Authorization check following permission revocation #### Bad Antipattern (AI Common Hallucination) > **Why it fails**: The write response's consistency context is discarded, and the subsequent check runs with default eventual consistency against potentially stale replicas, enabling unauthorized access immediately following revocation. ```typescript import { OpenFgaClient } from '@openfga/sdk'; export async function revokeAndVerifyAccess( fgaClient: OpenFgaClient, documentId: string, userId: string ): Promise<boolean> { // Anti-pattern: Tuple is deleted without capturing write token await fgaClient.write({ deletes: [{ user: `user:${userId}`, relation: 'viewer', object: `document:${documentId}`, }], }); // Flaw: Check uses default eventual consistency; // read replica may still return allowed=true due to replication lag (New Enemy bug) const response = await fgaClient.check({ user: `user:${userId}`, relation: 'viewer', object: `document:${documentId}`, }); return response.allowed ?? false; } ``` #### Production Standard (Correct) > **Standard**: By specifying ConsistencyPreference.HIGHER_CONSISTENCY, the check guarantees it evaluates against data at least as fresh as the latest commit, mitigating Zanzibar's New-Enemy race conditions. ```typescript import { OpenFgaClient, ConsistencyPreference } from '@openfga/sdk'; export interface RevokeResult { allowed: boolean; consistencyToken?: string; } export async function revokeAndVerifyAccess( fgaClient: OpenFgaClient, documentId: string, userId: string ): Promise<RevokeResult> { // Invariant: Capture write result containing consistency context const writeResult = await fgaClient.write({ deletes: [{ user: `user:${userId}`, relation: 'viewer', object: `document:${documentId}`, }], }); // Invariant: Explicitly require HIGHER_CONSISTENCY to prevent stale-replica reads const checkResponse = await fgaClient.check({ user: `user:${userId}`, relation: 'viewer', object: `document:${documentId}`, }, { consistency: ConsistencyPreference.HIGHER_CONSISTENCY, }); return { allowed: checkResponse.allowed ?? false, // Propagate transaction boundary to downstream services consistencyToken: writeResult.writes?.[0]?.tuple_key ? undefined : undefined, }; } ```

Architectural Specification & Invariant Rationale

Architectural Invariant Design: Mandatory ZedToken Causal Consistency Enforcement in OpenFGA ReBAC

Mandatory ZedToken Causal Consistency Enforcement in OpenFGA ReBAC defines an industrial-grade behavioral standard for security engineering. Distributed Relationship-Based Access Control (ReBAC) systems are vulnerable to Zanzibar's classic 'New Enemy' problem when mutations and authorization checks execute across asynchronously replicated data stores. Tuple writes modifying permissions must capture and propagate causal consistency tokens (ZedTokens or snapshot boundaries) to ensure downstream checks evaluate against at least as fresh a state snapshot. Omitting consistency constraints allows stale read-replica evaluations, resulting in critical authorization bypasses and privilege escalation after revocation events. 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: Executing unconstrained eventual-consistency authorization checks after relationship tuple mutations 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 Mode	AI Symptom (The Anti-Pattern)	Severity	Production Impact & Remedy
AI Hallucination & Deprecation Drift	Executing unconstrained eventual-consistency authorization checks after relationship tuple mutations	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 Inconsistency	AI 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:

Engineers MUST ALWAYS propagate the transaction consistency token returned from OpenFGA write operations (`writes` or `deletes`) through caller context or response headers.

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.