OpenFGA Transitive Relation Freshness and Cache Pruning Barrier

--- description: OpenFGA Transitive Relation Freshness and Cache Pruning Barrier - Guarantees zero-trust authorization consistency across high-velocity relationship tuple mutations and pre-computed evaluation caches. Mandates transactional cache pruning barriers and strict consistency token propagation to prevent authorization revocation lag and stale access leakage in transitive relation graphs. globs: src/**/*.{ts,tsx,js,jsx}, lib/**/*.{ts,tsx}, services/auth/**/*.{ts,tsx} alwaysApply: true verified_by: "agentrules.org" license: "CC-BY-4.0" source: "https://agentrules.org/rules/openfga-transitive-freshness-pruning-barrier" --- # OpenFGA Transitive Relation Freshness and Cache Pruning Barrier > **Severity**: CRITICAL | **Category**: security > **Anti-Pattern Warning**: Decoupled Fire-and-Forget Tuple Deletions Permitting Revocation Bypass via Stale Transitive Caches ## System Constraints You are a Principal Software Architect enforcing zero-trust authorization consistency in OpenFGA and Zanzibar-style authorization engines. You must enforce strict causal freshness, deterministic transitive cache pruning barriers, and consistency token propagation to prevent privilege escalation via stale relationship caches. ## Strict Guidelines - Engineers MUST ALWAYS enforce synchronous cache invalidation barriers or publish transactional outbox events when deleting or mutating relationship tuples in OpenFGA. - Engineers MUST NOT serve authorization checks from local or distributed pre-computed caches without asserting that the cache snapshot watermark is at least as fresh as the OpenFGA consistency token (Zookie/SnapToken). - Services MUST capture and propagate the OpenFGA `continuation_token` or consistency token returned from write operations to downstream callers and clients to guarantee read-after-write monotonic consistency. - Code SHOULD configure OpenFGA check evaluations with `consistency: ConsistencyPreference.HigherConsistency` for critical path revocations and high-risk administrative mutations. - Components NEVER assume that parent-child tuple deletion immediately collapses transitive evaluation nodes without explicitly invalidating memoized reachability subgraphs. ## Bad vs Good Implementations ### Tuple Revocation with Transitive Cache Invalidation Barrier and Freshness Token #### Bad Antipattern (AI Common Hallucination) > **Why it fails**: The tuple deletion executes without capturing the consistency token or purging downstream pre-computed caches. Distributed nodes evaluate subsequent authorization checks against stale cached reachability trees, enabling unauthorized access during the propagation lag window. ```typescript import { OpenFgaClient } from '@openfga/sdk'; const fgaClient = new OpenFgaClient({ apiUrl: process.env.FGA_API_URL, storeId: process.env.FGA_STORE_ID, }); // ANTI-PATTERN: Fire-and-forget deletion leaves transitive Redis cache intact // Clients reading immediately after revocation hit stale transitive caches export async function revokeGroupAccess(userId: string, groupId: string): Promise<void> { await fgaClient.write({ deletes: [ { user: `user:${userId}`, relation: 'member', object: `group:${groupId}`, }, ], }); // Transitive checks (e.g. document:123#viewer via group:eng) remain stale in cache } ``` #### Production Standard (Correct) > **Standard**: The good implementation synchronizes the tuple revocation with an atomic cache pruning barrier, records a freshness watermark, forces OpenFGA higher consistency verification, and returns the consistency token to guarantee read-after-write causal consistency across client boundaries. ```typescript import { OpenFgaClient, ConsistencyPreference } from '@openfga/sdk'; import { Redis } from 'ioredis'; interface RevocationResult { success: boolean; consistencyToken?: string; } export async function revokeGroupAccess( fgaClient: OpenFgaClient, redis: Redis, userId: string, groupId: string ): Promise<RevocationResult> { const userEntity = `user:${userId}`; const groupEntity = `group:${groupId}`; // 1. Transactionally mutate relationship tuple and obtain consistency watermark const writeResponse = await fgaClient.write({ deletes: [ { user: userEntity, relation: 'member', object: groupEntity, }, ], }); // 2. Pruning Barrier: Invalidate pre-computed reachability and evaluation cache tags const pipeline = redis.pipeline(); pipeline.del(`cache:authz:reachability:${userEntity}`); pipeline.srem(`cache:group:members:${groupEntity}`, userEntity); pipeline.set(`cache:authz:watermark:${userEntity}`, Date.now(), 'EX', 3600); await pipeline.exec(); // 3. Issue strong consistency check to barrier-sync OpenFGA replica nodes await fgaClient.check({ user: userEntity, relation: 'member', object: groupEntity, consistency: ConsistencyPreference.HigherConsistency, }); return { success: true, consistencyToken: writeResponse.continuation_token, }; } ```

Architectural Specification & Invariant Rationale

Architectural Invariant Design: OpenFGA Transitive Relation Freshness and Cache Pruning Barrier

OpenFGA Transitive Relation Freshness and Cache Pruning Barrier defines an industrial-grade behavioral standard for security engineering. Guarantees zero-trust authorization consistency across high-velocity relationship tuple mutations and pre-computed evaluation caches. Mandates transactional cache pruning barriers and strict consistency token propagation to prevent authorization revocation lag and stale access leakage in transitive relation graphs. 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: Decoupled Fire-and-Forget Tuple Deletions Permitting Revocation Bypass via Stale Transitive Caches 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	Decoupled Fire-and-Forget Tuple Deletions Permitting Revocation Bypass via Stale Transitive Caches	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 enforce synchronous cache invalidation barriers or publish transactional outbox events when deleting or mutating relationship tuples in OpenFGA.

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.