📦 @knolo/core

@knolo/core

KnoLo Core is a local-first knowledge base engine for small language models (LLMs).

It allows you to package structured documents into a compact .knolo file and query them deterministically — no embeddings, no vector databases, no cloud required.

Designed for:

• On-device LLMs
• Deterministic AI systems
• Agent routing
• Air-gapped or privacy-first environments

✨ Why KnoLo?

Traditional RAG systems require:

• Embeddings
• Vector databases
• External services
• Non-deterministic similarity scoring

KnoLo uses:

• Structured indexing
• Namespace-based routing
• Deterministic query resolution
• Compact .knolo bundles

This makes it:

• Fast
• Reproducible
• Lightweight
• Fully local

📦 Installation

npm install @knolo/core

🚀 Basic Usage

1️⃣ Mount a Knowledge Pack

import { mountPack } from "@knolo/core";

const pack = await mountPack("./dist/knowledge.knolo");

2️⃣ Query the Pack

import { query } from "@knolo/core";

const results = query(pack, {
  namespace: "mobile",
  q: "debounce vs throttle"
});

console.log(results);

3️⃣ Resolve an Agent

import { resolveAgent } from "@knolo/core";

const resolved = resolveAgent(pack, {
  agentId: "support-agent",
  query: "Explain debounce vs throttle",
});

Supports patch variables:

resolveAgent(pack, {
  agentId: "support-agent",
  patch: { tone: "formal" },
});

🤖 Agents

Agents are defined inside the pack metadata.

Phase 2 features include:

• Agent routing profiles
• Deterministic route validation
• Tool policies (allow_all, mixed, unknown)
• Registry validation at mount-time

🛠 Tool Policy Helpers

import { isToolAllowed, assertToolAllowed } from "@knolo/core";

isToolAllowed(agent, "web-search");
assertToolAllowed(agent, "database-read");

Default behavior:

• If no policy → allow all
• Explicit deny → deterministic error

📁 .knolo Format

A .knolo file contains:

• Indexed documents
• Namespaces
• Agent registry
• Metadata
• Routing profiles

Built using @knolo/cli.

🧠 Design Philosophy

KnoLo is built around:

• Determinism over probability
• Structure over embeddings
• Local-first AI
• Small model optimization
• Agent-native architecture

🔐 Use Cases

• On-device assistants
• Enterprise internal knowledge
• Mobile AI apps
• Secure environments
• Offline-first systems

🗺 Roadmap

• Rust core implementation
• WASM builds
• Multi-language SDKs
• Advanced agent routing
• Deterministic tool orchestration

📄 License

MIT