Quickstart: Node.js

1. Install

   npm install @equationalapplications/core-llm-wiki better-sqlite3

2. Create an adapter and the wiki

   memory.ts

   import Database from 'better-sqlite3';
   import OpenAI from 'openai';
   import { createWiki, type SQLiteAdapter } from '@equationalapplications/core-llm-wiki';
   
   const db = new Database('memory.db');
   
   const adapter: SQLiteAdapter = {
     async execAsync(sql) { db.exec(sql); },
     async runAsync(sql, params = []) {
       const info = db.prepare(sql).run(...(params as any[]));
       return { changes: info.changes, lastInsertRowId: Number(info.lastInsertRowid) };
     },
     async getAllAsync(sql, params = []) { return db.prepare(sql).all(...(params as any[])) as any; },
     async getFirstAsync(sql, params = []) { return (db.prepare(sql).get(...(params as any[])) ?? null) as any; },
     async withTransactionAsync(fn) {
       db.exec('BEGIN');
       try { const r = await fn(adapter); db.exec('COMMIT'); return r; }
       catch (e) { db.exec('ROLLBACK'); throw e; }
     },
     async closeAsync() { db.close(); },
   };
   
   const openai = new OpenAI();
   
   export const wiki = createWiki(adapter, {
     llmProvider: {
       generateText: async ({ systemPrompt, userPrompt }) => {
         const r = await openai.chat.completions.create({
           model: 'gpt-4o-mini',
           messages: [
             { role: 'system', content: systemPrompt },
             { role: 'user', content: userPrompt },
           ],
         });
         return r.choices[0].message.content ?? '{}';
       },
       embed: async (text) => {
         const r = await openai.embeddings.create({ model: 'text-embedding-3-small', input: text });
         return r.data[0].embedding;
       },
     },
   });
   
   await wiki.setup();

3. Ingest, read, and answer

   main.ts

   import { createHash } from 'node:crypto';
   import { readFile } from 'node:fs/promises';
   import { formatContext } from '@equationalapplications/core-llm-wiki';
   import { wiki } from './memory';
   
   const content = await readFile('handbook.md', 'utf8');
   const sourceHash = createHash('sha256').update(content).digest('hex');
   
   if (await wiki.hasChanged('team', 'handbook.md', sourceHash)) {
     await wiki.ingestDocument('team', { sourceRef: 'handbook.md', sourceHash, documentChunk: content });
   }
   
   const memory = await wiki.read('team', 'what is the on-call policy?');
   console.log(formatContext(memory));

Server checklist

• One connection per database file per process. The engine serializes its own transactions on that connection. Don’t open a second writer.
• Turning events into facts. Maintenance doesn’t run by itself: call runPendingMaintenance() on your own schedule, or set maintenance: 'auto' on a long-lived host. On a request-scoped or serverless host, always run it explicitly. See Serverless & multi-tenant hosting.
• Shutting down? In 'auto' maintenance mode, await wiki.drain() before you close, so background librarian and heal jobs can finish.