This specification is maintained by the AgentRules Working Group. Have enhancements or want to claim maintainership?Submit Specification →
backendcritical PriorityID: bun-sqlite-wal-snapshot-isolation-production-standard

Bun 1.2 Native SQLite WAL Concurrent Snapshot Isolation and Worker Visibility Standard

Enforces deterministic snapshot freshness and WAL checkpoint progression across concurrent Bun.Worker threads and asynchronous event loops using bun:sqlite. Eliminates stale read anomalies, WAL file bloat, and SQLITE_BUSY lock exhaustion by establishing explicit read transaction boundaries and isolated worker handles.

Prohibited Anti-Pattern: Dangling Read Snapshots and Cross-Worker Handle Contention in Bun SQLite WAL Mode

Export Ready-to-Drop Configuration

Select format below to copy code or download file directly.

---
description: Bun 1.2 Native SQLite WAL Concurrent Snapshot Isolation and Worker Visibility Standard - Enforces deterministic snapshot freshness and WAL checkpoint progression across concurrent Bun.Worker threads and asynchronous event loops using bun:sqlite. Eliminates stale read anomalies, WAL file bloat, and SQLITE_BUSY lock exhaustion by establishing explicit read transaction boundaries and isolated worker handles.
globs: src/**/*.ts, src/**/*.worker.ts
alwaysApply: true
verified_by: "agentrules.org"
license: "CC-BY-4.0"
source: "https://agentrules.org/rules/bun-sqlite-wal-snapshot-isolation-production-standard"
---

# Bun 1.2 Native SQLite WAL Concurrent Snapshot Isolation and Worker Visibility Standard

> **Severity**: CRITICAL | **Category**: backend
> **Anti-Pattern Warning**: Dangling Read Snapshots and Cross-Worker Handle Contention in Bun SQLite WAL Mode

## System Constraints
You are a Principal Software Architect enforcing Bun 1.2 Native SQLite WAL Concurrent Snapshot Isolation invariants. You must prevent stale read anomalies, SQLite lock starvation (SQLITE_BUSY), and WAL uncheck-pointed log bloat across concurrent worker pools and asynchronous event-loop microtasks. Enforce per-thread connection isolation, non-leaking prepared statement iterators, explicit transaction scopes, and strictly bounded PRAGMA WAL runtime configurations.

## Strict Guidelines
- Engineers MUST ALWAYS initialize Bun 1.2 `Database` instances with `PRAGMA journal_mode = WAL;`, `PRAGMA busy_timeout = 5000;`, and `PRAGMA synchronous = NORMAL;` to allow non-blocking concurrent readers alongside a serial writer.
- Engineers MUST NOT share a single `bun:sqlite` `Database` instance across concurrent `Bun.Worker` boundaries; each worker thread MUST open its own distinct database handle referencing the shared SQLite file.
- Engineers MUST ALWAYS exhaust or finalize query iterators and prepared statements within the same event-loop microtask tick to prevent long-lived read snapshots from blocking SQLite WAL checkpointing.
- Code SHOULD wrap sequential multi-query read operations requiring snapshot consistency in explicit read-only transactions (`BEGIN DEFERRED` / `COMMIT`) and immediately release them before initiating asynchronous I/O.
- Services MUST NOT execute unbounded `await` promises while holding open SQLite transactions or active prepared statement iterators, avoiding read-mark pinning and cross-worker write stalls.

## Bad vs Good Implementations
### Resolving Cross-Worker WAL Snapshot Lag and Read Lock Starvation

#### Bad Antipattern (AI Common Hallucination)
> **Why it fails**: Without WAL mode and busy_timeout, SQLite defaults to database-level locks. Suspending execution via `await` while prepared statements or implicit read locks remain active holds a read mark open in the SQLite WAL index. This causes cross-worker stale reads and prevents WAL checkpointing, eventually leading to catastrophic file bloat and SQLITE_BUSY write rejections.
```typescript
import { Database } from 'bun:sqlite';

// ANTI-PATTERN: Default journal mode (DELETE), no busy_timeout, holding statement open across await
export const db = new Database('production.db');

export async function getRecentOrdersAndAudit(userId: string) {
  const stmt = db.prepare('SELECT * FROM orders WHERE user_id = ?');
  // Unfinalized cursor holds open read snapshot across asynchronous network I/O
  const orders = stmt.all(userId);
  
  // External network hop pins the WAL read transaction indefinitely
  await fetch(`https://audit.internal.net/log?user=${userId}`);
  
  // Concurrent worker writes are invisible or trigger SQLITE_BUSY on checkpoint
  const count = db.query('SELECT COUNT(*) as total FROM orders WHERE user_id = ?').get(userId);
  return { orders, count };
}
```

#### Production Standard (Correct)
> **Standard**: Configures WAL mode, synchronous=NORMAL, and a busy_timeout. Encloses read queries in an explicit synchronous transaction function that terminates and releases the SQLite WAL read mark before any asynchronous `await` boundary. Creates isolated per-worker connections rather than sharing mutable handles.
```typescript
import { Database } from 'bun:sqlite';

export function createWorkerDatabaseConnection(): Database {
  const db = new Database('production.db', { strict: true });
  // Configure WAL invariants for non-blocking concurrent reads and safe checkpoints
  db.exec(`
    PRAGMA journal_mode = WAL;
    PRAGMA busy_timeout = 5000;
    PRAGMA synchronous = NORMAL;
  `);
  return db;
}

export async function getRecentOrdersAndAudit(userId: string): Promise<{ orders: unknown[]; count: number }> {
  const db = createWorkerDatabaseConnection();
  let orders: unknown[] = [];
  let count = 0;

  try {
    // Scope read snapshot strictly to synchronous transaction; release WAL mark immediately
    const readTx = db.transaction(() => {
      const ordersStmt = db.prepare('SELECT * FROM orders WHERE user_id = ?');
      const countStmt = db.prepare('SELECT COUNT(*) as total FROM orders WHERE user_id = ?');
      orders = ordersStmt.all(userId);
      const result = countStmt.get(userId) as { total: number } | null;
      count = result?.total ?? 0;
    });
    readTx(); // Read transaction completes and releases snapshot before async boundary
  } finally {
    db.close();
  }

  // Safe async I/O after SQLite lock/read-snapshot is completely relinquished
  await fetch(`https://audit.internal.net/log?user=${userId}`);
  return { orders, count };
}
```

Recommended MCP Servers (mcp-serverlist.com)

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

Recommended Agent Skills (askills.dev)

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

Architectural Specification & Invariant Rationale

Architectural Invariant Design: Bun 1.2 Native SQLite WAL Concurrent Snapshot Isolation and Worker Visibility Standard

Bun 1.2 Native SQLite WAL Concurrent Snapshot Isolation and Worker Visibility Standard defines an industrial-grade behavioral standard for backend engineering. Enforces deterministic snapshot freshness and WAL checkpoint progression across concurrent Bun.Worker threads and asynchronous event loops using bun:sqlite. Eliminates stale read anomalies, WAL file bloat, and SQLITE_BUSY lock exhaustion by establishing explicit read transaction boundaries and isolated worker handles. 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: Dangling Read Snapshots and Cross-Worker Handle Contention in Bun SQLite WAL Mode 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 DriftDangling Read Snapshots and Cross-Worker Handle Contention in Bun SQLite WAL ModeCritical
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 initialize Bun 1.2 `Database` instances with `PRAGMA journal_mode = WAL

Engineers MUST ALWAYS initialize Bun 1.2 `Database` instances with `PRAGMA journal_mode = WAL;`, `PRAGMA busy_timeout = 5000;`, and `PRAGMA synchronous = NORMAL;` to allow non-blocking concurrent readers alongside a serial writer.

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 share a single `bun

Engineers MUST NOT share a single `bun:sqlite` `Database` instance across concurrent `Bun.Worker` boundaries; each worker thread MUST open its own distinct database handle referencing the shared SQLite file.

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

Engineers MUST ALWAYS exhaust or finalize query iterators and prepared statements within the same event-loop microtask tick to prevent long-lived read snapshots from blocking SQLite WAL checkpointing.

Engineers MUST ALWAYS exhaust or finalize query iterators and prepared statements within the same event-loop microtask tick to prevent long-lived read snapshots from blocking SQLite WAL checkpointing.

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

Code SHOULD wrap sequential multi-query read operations requiring snapshot consistency in explicit read-only transactions (`BEGIN DEFERRED` / `COMMIT`) and immediately release them before initiating asynchronous I/O.

Code SHOULD wrap sequential multi-query read operations requiring snapshot consistency in explicit read-only transactions (`BEGIN DEFERRED` / `COMMIT`) and immediately release them before initiating asynchronous I/O.

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

Services MUST NOT execute unbounded `await` promises while holding open SQLite transactions or active prepared statement iterators, avoiding read-mark pinning and cross-worker write stalls.

Services MUST NOT execute unbounded `await` promises while holding open SQLite transactions or active prepared statement iterators, avoiding read-mark pinning and cross-worker write stalls.

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 Bun 1.2 Native SQLite WAL Concurrent Snapshot Isolation and Worker Visibility Standard critical for AI coding assistants?

Enforces deterministic snapshot freshness and WAL checkpoint progression across concurrent Bun.Worker threads and asynchronous event loops using bun:sqlite. Eliminates stale read anomalies, WAL file bloat, and SQLITE_BUSY lock exhaustion by establishing explicit read transaction boundaries and isolated worker handles. 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 (Dangling Read Snapshots and Cross-Worker Handle Contention in Bun SQLite WAL Mode) 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.

1 Rule in Stack

Bun