Quick Start

Welcome to PrimePilot! This guide will help you get started quickly by walking you through creating your first agent, adding knowledge sources, creating a squad, and deploying your AI assistant.

Step 1: Create Your Account

Begin by signing up for a PrimePilot account. Once you've created your account and logged in, you'll have access to the dashboard where you can manage agents, squads, and deployments. Your subscription plan determines which agent types and features are available to you. See Agent Catalog to understand what's included in each plan.

Important: After logging in, new users should follow the Setup Guide on the dashboard to complete initial configuration. This includes configuring storage, creating an API key, and adding a domain. These steps are essential for using PrimePilot's features effectively.

Step 2: Create Your First Agent

Navigate to the Agent Catalog in your dashboard to see the available agent types for your subscription. You can start with a simple proxy agent that connects directly to an LLM, or create a RAG agent that will be trained on your knowledge base. For most use cases, we recommend starting with a RAG agent as it provides the most value once you add knowledge sources. See What is an Agent? to learn about different agent types and their capabilities.

To create an agent, select an agent type from the catalog and click create. You'll be prompted to give your agent a name and optionally configure initial settings. An unconfigured RAG agent acts as a proxy agent until you add knowledge sources, so you can start using it immediately and enhance it later with your knowledge base.

Step 3: Add Knowledge Sources

If you created a RAG agent, the next step is to add knowledge sources to train it on your content. Navigate to the Knowledge Bases section and add your first knowledge source. You can upload documents in various formats, connect a website, add URLs, or import from other sources. PrimePilot automatically processes and indexes your content, creating embeddings that enable your agent to retrieve relevant information when answering questions.

The processing time depends on the size and type of your knowledge sources. Once processing is complete, your RAG agent will be able to answer questions based on the content you've added. You can continue adding more knowledge sources over time to expand your agent's knowledge base. See Knowledge Bases for information on supported knowledge source types and processing.

Step 4: Create a Squad

Once you have one or more agents configured, create a squad to orchestrate them together. Navigate to the Squad Catalog and create a new squad. Add your agents to the squad, keeping in mind that squads should be designed to serve a particular purpose. Select agents based on your use case, and consider mixing different types of agents to create a powerful solution. See Creating Squads for guidance on agent selection and squad design.

Configure your squad's orchestration settings to determine how agents work together. You can enable automatic agent switching based on intent detection, or allow manual agent switching during conversations. Squads with MCP agents become particularly powerful as they can integrate external tools and services. See What is a Squad? for details on squad orchestration capabilities.

Step 5: Deploy Your Squad

With your squad configured, you're ready to deploy it. PrimePilot offers multiple deployment options to suit different needs. Full Page Chat provides a full-page interface on the PrimePilot.ai domain, accessible via a unique URL or custom subdomain. Floating Widget adds a chat button to your website that opens a chat window. Inline Widget embeds a chat interface directly within your website content. API Integration allows programmatic access to your squad through RESTful endpoints. See the Deployment section for details on each deployment method and integration instructions.

Each deployment method provides configuration options for customization, including branding, positioning, and behavior settings. Copy the deployment configuration code provided for your chosen method and add it to your website or application. Your squad will be immediately available to users once deployed.

Next Steps

Now that you have your first agent and squad deployed, explore additional features to enhance your AI assistant. Configure your agents with custom system prompts and LLM settings, add more knowledge sources to expand capabilities, create additional squads for different use cases, monitor usage and analytics through the dashboard, and configure data privacy and retention policies. Continue learning by exploring the documentation sections on Agents, Squads, Deployment, and Admin Dashboard for advanced features and best practices.