aiknowledgedb

library2.1.4jsnpmunverified

Semantic knowledge database CLI for storing, searching, and listing knowledge entries using SQLite + sqlite-vec + FTS5. Current stable version is 2.1.4. Release cadence is irregular, with the last update in 2025. Key differentiators: lightweight local-first solution with embeddings support, optimized for semantic search with vector similarity, and full-text search via FTS5. Requires Node.js >=24.0.0. Designed as a CLI tool for developers working with knowledge management, especially in AI-assisted coding environments like Claude Code. Supports custom embedding models and SQLite-backed persistence.

database

npm install aiknowledgedb

github ↗package ↗homepage ↗

INSTALL

IMPORT

SIG · AIKNOWLEDGEDB

aiknowledgedb

databasejavascriptv2.1.4

harness data pending

Install & Compatibility

Where this runs

No compatibility data collected yet for this library.

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

cli

✓ import { cli } from 'aiknowledgedb'

✗ const cli = require('aiknowledgedb')

ESM-only package since v2; requires Node >=24. Check runtime compatibility. Use import syntax only.

KnowledgeBase

✓ import { KnowledgeBase } from 'aiknowledgedb'

Default export is not available; only named exports. TypeScript types are shipped with the package.

store

✓ import { store } from 'aiknowledgedb'

store is a top-level function for adding entries.

Initializes a KnowledgeBase, stores a sample entry, then performs a semantic search and logs results.

import { cli, KnowledgeBase, store, search } from 'aiknowledgedb'; const kb = new KnowledgeBase({ dbPath: './knowledge.db', embeddingModel: 'text-embedding-3-small', embeddingApiKey: process.env.OPENAI_API_KEY ?? '' }); await kb.init(); await store(kb, { title: 'Example Entry', content: 'This is a test knowledge entry.', tags: ['demo'] }); const results = await search(kb, { query: 'test entry', limit: 5 }); console.log(results); await kb.close();

Debug

Known issues

breakingBreaking change: Node.js 24 is required. Versions 1.x supported Node 18+.

fix

Upgrade Node.js to version >=24.0.0.

affects: >=2.0.0

breakingBreaking change: Package is now ESM-only in v2. CommonJS require() will fail.

fix

Use import syntax instead of require(). Ensure package.json contains "type": "module" if using in a CJS project.

affects: >=2.0.0

deprecated`search` function now returns Promise<Array<{id, title, content, score}>> instead of raw rows. Deprecated since v2.1.0.

fix

Update code to handle new result shape: { id, title, content, score }.

affects: >=2.1.0

gotchaEmbedding model API key must be set via environment variable or constructor option. If missing, initialisation fails silently.

fix

Always provide an apiKey via 'embeddingApiKey' option or set environment variable 'AIKNOWLEDGEDB_OPENAI_KEY'.

affects: >=2.0.0

gotchaDatabase file path is relative to current working directory. If not absolute, unexpected locations may be created.

fix

Use an absolute path for dbPath to avoid surprises: path.resolve('./knowledge.db').

affects: >=1.0.0

Errors

Common errors & fixes

Error [ERR_MODULE_NOT_FOUND]: Cannot find package 'aiknowledgedb' imported from ...

Package not installed or Node.js version is below 24.

fix

Run npm install aiknowledgedb@latest and ensure Node >=24. Use import instead of require.

TypeError: kb.init is not a function

Incorrect import; likely imported default instead of named export.

fix

Use import { KnowledgeBase } from 'aiknowledgedb' (not default import).

Error: OpenAI API key not configured

Missing embeddingApiKey option or environment variable.

fix

Set AIKNOWLEDGEDB_OPENAI_KEY environment variable or pass apiKey in constructor.

Upgrade

Version history

2.1.4latest on npm

Audit

Dependencies

No dependency data recorded yet.

Agent activity

36 hits · last 30 days

node

OpenAI (training)

Resources

githubhttps://github.com/mka-codelake/aiknowledgedb ↗

packageaiknowledgedb ↗

homepagehttps://github.com/mka-codelake/aiknowledgedb#readme ↗