Agent One LogoHelp Center
← Back to Homepage

Frequently Asked Questions

General Questions

What is Agent One?

Agent One is a platform for creating and managing AI-powered agents that can be integrated into websites or applications. Our agents can handle customer support, lead generation, data collection, and interact with external APIs.

How can Agent One benefit my project or business?

Agent One provides:

  • 24/7 automated customer support
  • Lead generation through email collection
  • Integration with your existing systems via custom APIs
  • Knowledge base integration for accurate responses
  • Multi-language support
  • Real-time analytics and insights

Account and Usage

How do I get started with Agent One?

  1. Sign up for an account
  2. Create a site (for public agents) or skip directly to agent creation (for private agents)
  3. Configure your agent's settings and knowledge base
  4. Deploy your agent

Can I use my own domain with Agent One?

Yes, you can either:

  • Use a custom domain that you own
  • Use a subdomain provided by Agent One (yoursite.agnt.one)
  • Embed the chat widget on any website

Creating and Managing Agents

How do I create an AI agent?

Follow our Creating an Agent guide for detailed steps. The basic process involves:

  1. Setting up basic configuration
  2. Adding knowledge base content
  3. Configuring tools and APIs
  4. Testing and deployment

What tools can my agent use?

Agents can use several built-in tools:

  • Code Interpreter for programming tasks
  • File Search for knowledge base access
  • Email Collection for lead generation
  • Custom API Functions for external integrations

How do I add custom API functionality to my agent?

  1. Go to your agent's settings
  2. Find the Custom API Functions section
  3. Click "Add API Function"
  4. Configure the endpoint, parameters, and description
  5. Test the integration before deployment

How does email collection work?

You can configure your agent to collect emails:

  • Choose between collecting at the start or end of conversations
  • The agent will naturally request the email
  • Collected emails are stored securely in your dashboard

Technical Questions

What types of APIs can I integrate with my agent?

Your agent can integrate with:

  • RESTful APIs
  • Public web services
  • Internal APIs (must be publicly accessible)
  • APIs with URL parameters

Note: Currently, only GET requests are supported, and APIs should not require authentication.

How do I format API parameters?

Use JSON Schema format to define parameters. Example:

{
  "parameter_name": {
    "type": "string",
    "description": "What this parameter does",
    "required": true
  }
}

Can I use URL parameters in my API endpoints?

Yes, use curly braces for parameters: /api/weather/{city}. The agent will automatically replace these with actual values during calls.

Is there a limit to how many custom APIs I can add?

While there's no strict limit, we recommend keeping the number of custom APIs manageable for better agent performance and easier maintenance.

Privacy and Security

How secure are the API integrations?

  • All API calls are made server-side
  • Only public endpoints are supported
  • HTTPS is required for all integrations
  • API responses are securely handled

How is collected email data stored?

  • Emails are stored securely in our database
  • Access is limited to authorized team members

Support and Troubleshooting

What should I do if my custom API isn't working?

  1. Verify the API endpoint is publicly accessible
  2. Check the parameter configuration
  3. Test the API independently
  4. Review the error messages in your dashboard
  5. Contact support if issues persist

How can I get help if I'm having issues?

Please contact our support team through the chat widget on the bottom right of the homepage.

Agent Customization

Can I deploy multiple agents on the same site?

Yes, you can create and manage multiple agents for a single site:

  1. Navigate to your site's Agents page
  2. Create new agents as needed
  3. Each agent can have different configurations and purposes
  4. You can embed different agents on different pages of your website
  5. Your main subdomain page will show a list of all your agents

How do I view analytics for my agent?

To access analytics:

  1. Go to your dashboard
  2. Select your site
  3. Click on "Analytics" in the left navigation
  4. View metrics including:
    • Pageviews
    • Unique visitors
    • Top pages
    • Referrer sources
    • Visitor countries
    • Bounce rates

Analytics data is updated hourly and available for the last 30 days.

Website Integration

How do I get the embed code for my agent?

To get your embed code:

  1. Go to your agent's editor page
  2. Look for the "Embed Code" button - You will need to create a site first, private agents do not have the embed code option
  3. Click it to open a dialog with your custom code
  4. Use the "Copy to Clipboard" button to copy the code
  5. Paste it into your website's HTML just before the closing </body> tag

Will the chat widget work with my existing website design?

Yes, the chat widget is designed to work with virtually any website:

  • It appears as a small button in the corner of your site
  • When clicked, it expands into a chat interface
  • It adapts to light/dark mode based on user preferences
  • The widget is responsive and works on mobile devices

Can users upload files to my agent?

Currently, file upload capabilities are limited to specific subscription tiers. This feature may not be available on all plans.

Model Capabilities

What's the difference between the available AI models?

Agent One offers different models with varying capabilities: Choose the model that best fits your specific use case and performance requirements.

Can my agent understand and respond in multiple languages?

Yes, your agent can handle multiple languages. For best results:

  1. Specify supported languages in your agent instructions
  2. Include examples in those languages if possible
  3. Upload knowledge base documents in the relevant languages
  4. Test your agent with queries in different languages

How can I make my agent's responses more accurate?

To improve accuracy:

  1. Provide detailed instructions in the agent configuration
  2. Include specific examples of ideal responses
  3. Upload comprehensive and well-structured knowledge base documents
  4. Use the Preview feature to test and refine responses
  5. Review actual conversations and update your configuration based on performance