Everything you need to build with Aurora Vault

Comprehensive guides, API references, and integration tutorials — all in one place.

Quick Start Guide

Get your Aurora Vault workspace running in under 5 minutes. This guide walks you through account creation, workspace configuration, and your first automation.

1. Create your account

Visit app.auroravault.ai and sign up with your work email. You'll be placed in a 14-day full-access trial on the Team plan automatically — no credit card required.

2. Configure your workspace

Choose your workspace name, invite team members (or skip for solo), and select your primary use case. Aurora will pre-configure your dashboard layout based on your role.

3. Connect your first integration

Navigate to Settings → Integrations and connect your first app. We recommend starting with the tool you use most — Slack, Gmail, or Notion are popular first picks.

// Initialize Aurora Vault SDK import AuroraVault from '@aurora/sdk'; const vault = new AuroraVault({ apiKey: process.env.AURORA_API_KEY, workspace: 'your-workspace-id' }); // Create your first automated task const task = await vault.tasks.create({ title: 'Generate weekly summary', agent: 'aurora-assistant', schedule: 'every Monday 9:00am' });

4. Run your first AI task

In the dashboard, click New Task and describe what you want to accomplish in plain English. Aurora's AI will interpret your intent and confirm the execution plan before running.

Workflow Builder

The Aurora Workflow Builder lets you create multi-step automations with a visual, no-code interface. Add triggers, actions, conditions, and AI logic nodes in any order.

Workflow Structure

Every workflow has three core components: a Trigger (what starts it), Steps (what happens), and Actions (where outputs go).

# Aurora Workflow Configuration (YAML) workflow: name: "Weekly Report Generator" trigger: type: schedule cron: "0 9 * * 1" # Every Monday 9am steps: - id: gather_data action: aurora.knowledge.query params: query: "Team performance last 7 days" - id: generate_report action: aurora.ai.generate params: prompt: "Create executive summary from: {{gather_data.result}}" tone: professional - id: distribute action: slack.message.send params: channel: "#leadership" content: "{{generate_report.output}}"

AI Logic Nodes

Insert AI decision points between steps to add intelligence. Aurora can analyze step outputs, make conditional decisions, and route workflows based on semantic understanding — not just rule matching.

API Authentication

All Aurora Vault API calls are authenticated using Bearer tokens. Generate your API key from Settings → Developer → API Keys.

Request Headers

# Required headers for all API requests Authorization: Bearer av_live_xxxxxxxxxxxxxxxxxxxx Content-Type: application/json X-Workspace-ID: ws_xxxxxxxxxxxxxxxx

Rate Limits

API rate limits vary by plan. Solo plans allow 100 requests/minute, Team plans allow 1,000/minute, and Enterprise plans receive custom limits. Rate limit headers are included in every response.

# Rate limit response headers X-RateLimit-Limit: 1000 X-RateLimit-Remaining: 847 X-RateLimit-Reset: 1704067200

AI Agents

Aurora Agents are persistent AI workers that run in the background, monitor conditions, and take actions on your behalf. Each agent has a defined purpose, toolset, and memory scope.

Creating an Agent

Agents are configured through the Agents panel in your workspace, or programmatically via the API. Define their instructions in plain language, then assign tools they're allowed to use.

// Create a persistent AI agent const agent = await vault.agents.create({ name: 'Inbox Manager', instructions: ` Monitor incoming emails. Categorize each as: - Urgent: Respond within 1 hour - Standard: Draft a reply and queue for review - Low priority: Archive and summarize weekly `, tools: ['gmail.read', 'gmail.draft', 'aurora.summarize'], memory: 'persistent', autoApprove: false // Require human review before sending });

Agent Safety

All agents operate in sandboxed environments with explicit tool permissions. You control exactly what each agent can and cannot do. Critical actions (like sending emails or making API calls) can be gated behind human approval workflows.