Metalogue
Docs

Getting Started

Get up and running with the Metalogue SDK in 5 minutes.

Getting Started

Get up and running with Metalogue in 5 minutes.

Prerequisites

  • Node.js 18+ / Python 3.9+ / Go 1.21+
  • A Metalogue API key (get one here)

Installation

Choose your language:

TypeScript / Node.js

npm install @metalogue/sdk
# or
yarn add @metalogue/sdk
# or
pnpm add @metalogue/sdk

Python

pip install metalogue
# or
poetry add metalogue

Go

go get github.com/metalogue/metalogue-sdk-go

Quick Start

1. Initialize the Client

TypeScript:

import { MetalogueClient } from '@metalogue/sdk';

const client = new MetalogueClient({
  apiKey: process.env.METALOGUE_API_KEY,
});

Python:

from metalogue import MetalogueClient

client = MetalogueClient(api_key=os.environ["METALOGUE_API_KEY"])

Go:

import metalogue "github.com/metalogue/metalogue-sdk-go"

client := metalogue.NewClient(os.Getenv("METALOGUE_API_KEY"))

2. Connect a Data Source

Add your first connector (e.g., Slack):

const connector = await client.createConnector({
  connector_type: 'slack',
  display_name: 'Company Slack',
  credentials: { token: 'xoxb-your-slack-token' },
});

// Trigger initial sync
await client.syncConnector(connector.connector_id, true);

3. Query Your Data

Once synced, query across all connected sources:

const { results, reasoning, latencyMs } = await client.query(
  'What was the decision on authentication architecture?',
  {
    limit: 10,
    includeReasoning: true,
  }
);

console.log(`Found ${results.length} results in ${latencyMs}ms`);

for (const result of results) {
  console.log(`[${result.source}] ${result.content}`);
  console.log(`  Score: ${result.score}`);
  console.log(`  URL: ${result.url}`);
}

Core Concepts

Connectors

Connectors are data sources you connect to Metalogue. We support 41+ tools:

  • Productivity: Slack, Notion, Google Workspace, Microsoft 365
  • DevTools: GitHub, GitLab, Jira, Linear
  • CRM: Salesforce, HubSpot, Pipedrive
  • Support: Zendesk, Freshdesk, Intercom

Each connector syncs documents into the federated graph.

Federated Query

Queries search across all connected sources simultaneously:

const response = await client.query('quarterly report', {
  sources: ['google-drive', 'notion'],  // Optional: filter sources
  dateRange: {
    after: '2025-01-01',
    before: '2025-12-31',
  },
});

Ephemeral Context (Ghost Graph)

Results are assembled into an Ephemeral Context—a just-in-time relationship graph that:

  • Connects related documents
  • Respects token limits
  • Enables multi-hop reasoning

Intent Ranking

Results are ranked using 5D intent vectors:

  • Urgency - How time-sensitive is this?
  • Emotional Weight - Rational vs emotional importance
  • Action Likelihood - Browsing vs intent to act
  • Social Dimension - Solo vs collaborative context
  • Context Sensitivity - How context-dependent

Next Steps

Now that you're set up:

  1. Connect more sources - Add Notion, GitHub, and more
  2. Explore the API - See Sync, Query, Bridge
  3. Set up compliance - Configure Forget Request
  4. Deploy to production - Use our SDKs in your apps

Getting Help