How can we help?

Getting your first Alfred AI agent up and running takes just a few minutes. Follow these steps to create a fully functional AI agent.

Step 1: Access the Dashboard

Log in to your account at gositeme.com/dashboard. If you don't have an account yet, sign up for a plan first.

Step 2: Create a New Agent

1. Click the "+ New Agent" button in the top-right corner of your dashboard.
2. Choose a name for your agent (e.g., "Customer Support Bot").
3. Select a base template or start from scratch.
4. Choose your agent's primary language.

Step 3: Configure Your Agent

Set the agent's personality, tone, and knowledge base. You can upload documents, link URLs, or type in custom instructions.

Step 4: Test & Deploy

Use the built-in chat preview to test your agent. When you're satisfied, click "Deploy" to make it live. You can embed it on your website, connect it to a phone number, or integrate via API.

For more, see our Getting Started documentation.

The Alfred AI dashboard is your command center for managing agents, monitoring performance, and configuring settings.

Dashboard Layout

• Left Sidebar: Navigation to Agents, Fleet, Voice, Analytics, Settings, and more.
• Top Bar: Quick search, notifications, and account menu.
• Main Area: Content panels that change based on where you navigate.
• Status Bar: Current plan, usage meters, and quick actions.

Key Sections

Agents Panel: View, create, edit, and delete your AI agents. Each card shows status (active/paused), conversation count, and satisfaction score.

Analytics: Track usage, conversations, token consumption, and customer satisfaction in real time. Access via Analytics.

Quick Actions: Use the Ctrl+K (or ⌘+K) shortcut to open the command palette for rapid navigation.

Alfred AI supports voice interactions out of the box. Here's how to set up voice for your agent.

Step 1: Get a Phone Number

Navigate to Voice & AI Products and purchase a phone number. Numbers are available in 60+ countries.

Step 2: Assign to an Agent

1. Go to your agent's settings in the Dashboard.
2. Click the "Voice" tab.
3. Select your purchased phone number from the dropdown.
4. Choose a voice (default or clone your own).

Step 3: Configure Voice Settings

Set greeting message, call timeout, transfer number for escalation, and working hours. You can also enable voicemail and call recording.

Step 4: Test It

Call your assigned number to test the experience. Use the Voice Portal to monitor live calls.

Alfred AI agents can be deployed across multiple channels to meet your customers wherever they are.

Available Channels

• Web Widget: Embed on any website with a single script tag.
• Phone / Voice: Assign a phone number for voice interactions.
• API: Connect via REST API for custom integrations.
• Webhooks: Receive real-time events in your applications.
• White-Label: Fully branded experience for your customers — see White-Label.

Embedding the Web Widget

<script src="https://gositeme.com/assets/js/alfred-widget.js"
  data-agent-id="YOUR_AGENT_ID"
  data-theme="dark"></script>

Place this code before the closing </body> tag on your website. The widget will appear in the bottom-right corner by default.

Alfred AI provides a powerful REST API. Here's how to make your first API call.

Step 1: Get Your API Key

Go to Developer Portal → API Keys → "Create New Key". Copy and store your key securely — it won't be shown again.

Step 2: Make a Request

curl -X POST https://gositeme.com/api/v1/chat \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"agent_id": "ag_abc123", "message": "Hello!"}'

Step 3: Handle the Response

{
  "status": "success",
  "response": "Hello! How can I help you today?",
  "conversation_id": "conv_xyz789",
  "tokens_used": 42
}

For full API documentation, visit the API Reference.

You can manage your Alfred AI subscription at any time from your account settings.

Viewing Your Current Plan

Go to Dashboard → Settings → Subscription. Here you'll see your current plan, renewal date, and usage summary.

Changing Your Plan

1. Click "Change Plan" from the subscription page.
2. Compare available plans on the Pricing page.
3. Select your desired plan and confirm.
4. If upgrading, the prorated difference is charged immediately.
5. If downgrading, the change takes effect at your next billing cycle.

Cancellation

To cancel, go to Settings → Subscription → "Cancel Subscription". Your account will remain active until the end of your current billing period. All data is retained for 30 days after cancellation.

Each Alfred AI plan includes specific usage allocations. Understanding these limits helps you choose the right plan.

What Counts as Usage?

• Tokens: Every AI interaction consumes tokens. A typical conversation uses 500–2,000 tokens.
• Conversations: A conversation is a complete session from start to finish.
• Voice Minutes: Inbound and outbound call minutes are tracked separately.
• Agents: Number of active agents you can run simultaneously.

Monitoring Usage

Track your usage in real time at Analytics. The dashboard shows daily/monthly trends with visual progress bars.

What Happens at the Limit?

When you reach 80% of your allocation, you'll receive an email notification. At 100%, agents pause until the next billing cycle unless you purchase a Token Pack add-on.

Upgrading unlocks more agents, higher token limits, advanced features, and priority support.

How to Upgrade

1. Visit the Pricing page or go to Dashboard → Settings → Subscription.
2. Click "Upgrade" next to your desired plan.
3. Review the feature comparison.
4. Confirm payment — you'll only be charged the prorated amount for the remainder of your current billing cycle.

Plan Comparison

• Starter ($3.99/mo): 1 agent, 10K tokens, basic tools.
• Pro ($19.99/mo): 5 agents, 100K tokens, voice, analytics.
• Business ($49.99/mo): 25 agents, 500K tokens, fleet management, API access.
• Enterprise (Custom): Unlimited agents, dedicated support, SLA, SSO, RBAC.

See full details on the Compare page.

Alfred AI accepts multiple payment methods for your convenience.

Accepted Payment Methods

• Credit/Debit Cards: Visa, Mastercard, American Express, Discover.
• PayPal: Link your PayPal account for automatic billing.
• Bank Transfer: Available for Enterprise plans (annual billing only).
• Crypto: Bitcoin and Ethereum accepted for annual plans.

Updating Your Payment Method

1. Go to Dashboard → Settings → Billing.
2. Click "Update Payment Method".
3. Enter your new card details or link PayPal.
4. Click "Save".

Payment Security

All payments are processed through Stripe with PCI-DSS Level 1 compliance. We never store your full card number on our servers.

Access and download invoices for all your Alfred AI payments.

Viewing Invoices

1. Go to Dashboard → Settings → Billing → "Invoice History".
2. You'll see a list of all past invoices with date, amount, and status.
3. Click any invoice to view details or download a PDF.

Invoice Contents

Each invoice includes: plan name, billing period, usage breakdown, taxes, total amount, and payment method used.

Tax Information

If you need to add a VAT/Tax ID to your invoices:

1. Go to Settings → Billing → "Tax Information".
2. Enter your company name and tax number.
3. All future invoices will include this information. You can also regenerate past invoices.

Alfred AI lets you build custom agents tailored to your specific business needs. Here's a complete guide.

Agent Configuration Options

• Name & Avatar: Give your agent a memorable identity.
• System Prompt: Define the agent's personality, tone, and boundaries.
• Knowledge Base: Upload documents (PDF, DOCX, TXT) or provide URLs for the agent to learn from.
• Tools: Enable specific tools the agent can use (web search, calculations, etc.).
• Behavior Rules: Set guardrails for topics the agent should avoid.

Writing Effective System Prompts

A good system prompt includes:

1. Role: "You are a customer support specialist for [Company]."
2. Tone: "Be friendly, professional, and concise."
3. Knowledge scope: "Only answer questions about our products."
4. Escalation: "If unsure, offer to transfer to a human agent."

Agent templates are pre-configured agents you can deploy instantly and customize to your needs.

Finding Templates

Browse templates at Agent Templates or from your dashboard under "New Agent" → "From Template".

Available Template Categories

• Customer Support: Handle FAQs, troubleshoot issues, process returns.
• Sales: Qualify leads, book demos, answer product questions.
• Healthcare: Appointment scheduling, patient intake, symptom triage.
• Real Estate: Property inquiries, showing scheduling, mortgage calculators.
• Legal: Initial consultations, document preparation, case intake.

Customizing a Template

1. Click "Use Template" on your chosen template.
2. The template creates a new agent with all pre-configured settings.
3. Edit the system prompt, tools, and knowledge base to match your business.
4. Test and deploy when ready.

Fleets allow you to coordinate multiple agents working together as a team. This is ideal for complex workflows.

What Is a Fleet?

A fleet is a group of agents assigned roles that collaborate to handle tasks. For example, a support fleet might have a triage agent, a technical agent, and a billing agent.

Creating a Fleet

1. Go to Fleet Dashboard.
2. Click "Create Fleet".
3. Name your fleet and define its purpose.
4. Add agents and assign roles (Leader, Specialist, Support).
5. Configure routing rules for how conversations are distributed.

Fleet Routing

• Intent-Based: Route based on detected customer intent.
• Round-Robin: Distribute evenly across agents.
• Priority: Route to the best-matched agent first.
• Escalation Chain: Cascade to next agent if first can't resolve.

Each agent in a fleet can be assigned specific roles and tasks to optimize performance.

Built-in Roles

• Leader: Receives all incoming conversations first, triages and delegates to specialists.
• Specialist: Handles specific topic areas (billing, technical, sales).
• Support: Assists other agents and handles overflow.
• Observer: Monitors conversations for quality assurance without interacting.

Assigning Tasks

Tasks define what an agent can do beyond conversation:

• Send emails and follow-up messages
• Create support tickets
• Update CRM records
• Schedule appointments
• Process payments

Task Configuration

In your agent's settings, go to "Tasks" tab → "Add Task". Select from available task types and configure the required parameters (API endpoints, credentials, templates).

Track how your agents are performing with Alfred AI's built-in analytics and monitoring tools.

Key Metrics

• Response Time: Average time to first response.
• Resolution Rate: Percentage of conversations resolved without escalation.
• Satisfaction Score: Based on user thumbs-up/down feedback.
• Token Efficiency: Tokens used per conversation (lower is better).
• Conversation Volume: Total interactions over time.

Accessing Analytics

Navigate to Analytics for platform-wide metrics, or click an individual agent to see agent-specific data.

Setting Up Alerts

1. Go to Settings → Notifications → "Performance Alerts".
2. Set thresholds (e.g., alert if satisfaction drops below 80%).
3. Choose notification channel: email, SMS, or webhook.

Voice agents handle phone calls automatically with natural-sounding AI conversations.

Prerequisites

• An active Alfred AI account (Pro plan or higher).
• A purchased phone number from Voice Products.
• At least one configured agent.

Setup Steps

1. Go to Voice Portal.
2. Click "New Voice Agent".
3. Select the AI agent that will handle calls.
4. Assign your phone number.
5. Choose a voice: select from 30+ premade voices or clone your own.
6. Set the greeting message and call flow.
7. Configure business hours and after-hours behavior.

Voice Settings

• Speech Speed: Adjust how fast the agent speaks (0.5x–2x).
• Silence Timeout: How long to wait for caller response (default: 5 seconds).
• Max Call Duration: Set a maximum call length.
• Transfer Number: Number to forward to if caller requests a human.

The IVR Builder lets you create interactive voice menus with a drag-and-drop visual editor.

What Is an IVR?

An Interactive Voice Response (IVR) system greets callers and routes them through menu options using voice or keypad input. Example: "Press 1 for Sales, Press 2 for Support."

Creating an IVR Flow

1. Open the IVR Builder.
2. Drag a "Greeting" node onto the canvas.
3. Add "Menu" nodes with options (1–9, or voice keywords).
4. Connect nodes to actions: transfer to agent, play message, collect input, or hang up.
5. Set fallback behavior for unrecognized inputs.

Advanced Features

• AI-Powered Routing: Let Alfred understand natural language instead of rigid menus.
• Time-Based Routing: Different flows for business hours vs. after-hours.
• Data Collection: Gather caller info (account number, case ID) before routing.
• CRM Integration: Look up caller in your CRM for personalized greetings.

Voice Cloning lets you create a custom AI voice that sounds like a specific person — great for brand consistency.

How It Works

1. Record or upload a voice sample (minimum 30 seconds, recommended 3+ minutes).
2. Our AI analyzes pitch, tone, cadence, and pronunciation patterns.
3. A custom voice model is generated (takes 5–15 minutes).
4. Assign the cloned voice to any of your agents.

Recording Tips

• Use a quiet room with minimal echo.
• Speak naturally at your normal pace.
• Read varied content (not just one sentence repeated).
• Use a good quality microphone (USB condenser recommended).

Legal Requirements

You must have explicit consent from the person whose voice is being cloned. When creating a voice clone, you'll need to confirm:

• You are the person, or have written consent from them.
• The voice will be used for legitimate business purposes.
• You agree to our voice cloning terms of service.

Call Campaigns let you make automated outbound calls at scale — perfect for appointments, reminders, surveys, and sales outreach.

Creating a Campaign

1. Go to Call Campaigns.
2. Click "New Campaign".
3. Name your campaign and select the agent that will handle calls.
4. Upload a contact list (CSV with name, phone, optional custom fields).
5. Set the call script — the agent's opening message and conversation flow.
6. Configure calling hours and timezone rules.

Campaign Settings

• Concurrency: Number of simultaneous calls (1–50).
• Retry Logic: How many times to retry unanswered calls.
• Caller ID: Which number appears on the recipient's phone.
• Compliance: Auto-skip numbers on do-not-call lists.

Monitoring Results

Track campaign progress in real time: calls completed, answered, voicemail left, outcomes, and conversion rates.

Conference Rooms enable multi-party calls with AI agents — great for meetings where AI assists in real time.

Use Cases

• AI-Assisted Meetings: An agent joins your conference to take notes, summarize decisions, and create action items.
• Training Sessions: Multiple new users learn with an AI trainer on the call.
• Customer Calls: Your team + AI agent collaborate on complex customer issues.
• Interviews: AI assists with question prompts and candidate evaluation.

Setting Up a Conference Room

1. Go to Conference Rooms.
2. Click "Create Room".
3. Set the room name, PIN (optional), and max participants.
4. Choose which AI agent(s) to include.
5. Share the dial-in number and PIN with participants.

During the Conference

The AI agent can be configured to:

• Listen silently and generate a summary afterward.
• Actively participate and answer questions.
• Provide real-time fact-checking and data lookups.

Alfred AI supports two authentication methods: API Keys (for server-to-server) and OAuth 2.0 (for user-authorized applications).

API Key Authentication

1. Go to Developer Portal → API Keys.
2. Click "Create New Key".
3. Name your key (e.g., "Production Server") and set permissions.
4. Copy the key immediately — it's shown only once.

Use the key in requests:

Authorization: Bearer sk_live_abc123def456

OAuth 2.0

For applications that act on behalf of users:

1. Register your app in the Developer Portal → OAuth Apps.
2. Set your redirect URI(s).
3. Use the authorization code flow to obtain access tokens.

GET /api/oauth/authorize?client_id=YOUR_ID&redirect_uri=YOUR_URI&response_type=code&scope=agents:read+agents:write

Key Security Best Practices

• Rotate keys every 90 days.
• Use environment variables — never hardcode keys.
• Set minimum required permissions for each key.
• Use separate keys for development and production.

The Alfred AI REST API lets you programmatically manage agents, conversations, and voice features.

Base URL

https://gositeme.com/api/v1/

Common Endpoints

Example: List Agents

curl https://gositeme.com/api/v1/agents \
  -H "Authorization: Bearer YOUR_API_KEY"

Response Format

All responses are JSON with this structure:

{
  "status": "success",
  "data": { ... },
  "meta": { "page": 1, "total": 42 }
}

For complete API documentation, visit the API Reference.

Official SDKs make integrating Alfred AI into your application faster and easier. Available for three languages.

Node.js

npm install @gositeme/alfred-sdk

const Alfred = require('@gositeme/alfred-sdk');
const client = new Alfred({ apiKey: process.env.ALFRED_API_KEY });

const response = await client.chat.send({
  agentId: 'ag_abc123',
  message: 'Hello!'
});

Python

pip install gositeme-alfred

from gositeme_alfred import Alfred

client = Alfred(api_key=os.environ['ALFRED_API_KEY'])
response = client.chat.send(
    agent_id='ag_abc123',
    message='Hello!'
)

PHP

composer require gositeme/alfred-sdk

use GoSiteMe\Alfred\Client;

$client = new Client(getenv('ALFRED_API_KEY'));
$response = $client->chat->send([
    'agent_id' => 'ag_abc123',
    'message'  => 'Hello!'
]);

Webhooks let you receive real-time notifications when events occur in your Alfred AI account.

Setting Up Webhooks

1. Go to Developer Portal → Webhooks.
2. Click "Add Endpoint".
3. Enter your HTTPS callback URL.
4. Select the events you want to subscribe to.
5. Save and copy the webhook secret for signature verification.

Available Events

• conversation.started — New conversation began
• conversation.ended — Conversation completed
• message.received — New message in a conversation
• agent.status_changed — Agent went online/offline
• call.started / call.ended — Voice call events
• usage.threshold — Usage limit approaching

Verifying Webhook Signatures

const crypto = require('crypto');
const signature = req.headers['x-alfred-signature'];
const expected = crypto
  .createHmac('sha256', WEBHOOK_SECRET)
  .update(req.body)
  .digest('hex');
const valid = crypto.timingSafeEqual(
  Buffer.from(signature), Buffer.from(expected)
);

The Alfred AI API enforces rate limits to ensure fair usage and platform stability.

Rate Limits by Plan

Rate Limit Headers

Every API response includes these headers:

X-RateLimit-Limit: 300
X-RateLimit-Remaining: 287
X-RateLimit-Reset: 1709510460

Common Error Codes

Handling Rate Limits

async function apiCall(url, options, retries = 3) {
  const res = await fetch(url, options);
  if (res.status === 429 && retries > 0) {
    const wait = parseInt(res.headers.get('Retry-After') || '5');
    await new Promise(r => setTimeout(r, wait * 1000));
    return apiCall(url, options, retries - 1);
  }
  return res.json();
}

Enterprise organizations get a dedicated workspace with advanced management features.

Initial Setup

1. After signing your Enterprise agreement, you'll receive an activation email.
2. Click the activation link to set up your organization.
3. Configure: organization name, logo, primary domain, and admin account.
4. Set up billing (invoice or credit card).

Organization Settings

Access at Enterprise Admin:

• General: Name, logo, contact info, and default timezone.
• Security: Password policies, 2FA requirements, IP allowlists.
• Compliance: Data retention policies, export controls, audit settings.
• Integrations: SSO, SCIM provisioning, API access controls.

Multi-Environment Support

Enterprise plans include separate environments:

• Production: Live agents serving customers.
• Staging: For testing changes before going live.
• Development: For building and experimenting.

Role-Based Access Control (RBAC) lets you define exactly what each team member can see and do.

Managing Team Members

1. Go to Enterprise Admin → Team.
2. Click "Invite Member".
3. Enter their email and assign a role.
4. They'll receive an invitation to join your organization.

Built-in Roles

Custom Roles

Create custom roles with granular permissions:

1. Go to Team → "Manage Roles" → "Create Custom Role".
2. Name the role and toggle individual permissions (agents, analytics, voice, billing, API, admin).
3. Assign to team members as needed.

Enterprise plans support Single Sign-On (SSO) via SAML 2.0, allowing your team to log in with your existing identity provider.

Supported Identity Providers

• Okta
• Azure Active Directory
• Google Workspace
• OneLogin
• Any SAML 2.0-compatible provider

Setup Steps

1. Go to Enterprise Admin → Security → "SSO Configuration".
2. Select your identity provider or choose "Custom SAML".
3. Enter the SSO URL, Entity ID, and upload the X.509 certificate from your IdP.
4. Configure attribute mappings (email, name, role).
5. Set the ACS (Assertion Consumer Service) URL in your IdP: https://gositeme.com/api/oauth/saml/callback
6. Test the connection before enabling for all users.

SSO Policies

• Enforce SSO: Require all members to log in via SSO (disable password login).
• Auto-Provisioning: Automatically create accounts for new SSO users.
• Role Mapping: Map IdP groups to Alfred AI roles.

White-Label lets you rebrand Alfred AI completely and offer it to your own customers under your brand.

What You Can Customize

• Brand: Logo, colors, fonts, favicon.
• Domain: Use your own domain (e.g., ai.yourcompany.com).
• Email: Send notifications from your domain.
• Interface: Custom dashboard layout and terminology.
• Documentation: Branded docs and help pages.

Setup Steps

1. Go to Enterprise Admin → White-Label.
2. Upload your logo (SVG or PNG, 200×50 recommended).
3. Set your brand colors (primary, secondary, accent).
4. Add your custom domain and configure DNS (CNAME record).
5. Customize email templates with your branding.
6. Set up your billing (you set pricing for your customers).

DNS Configuration

CNAME  ai.yourcompany.com  →  whitelabel.gositeme.com

SSL is automatically provisioned for your custom domain.

Audit logging provides a complete record of all actions taken in your organization for compliance and security purposes.

What's Logged

• Authentication: Logins, logouts, failed attempts, SSO events.
• Agent Changes: Created, updated, deleted, deployed agents.
• Team Changes: Members added/removed, role changes.
• API Activity: Key creation, key usage, OAuth grants.
• Settings: Configuration changes, billing updates.
• Data Access: Conversation exports, report downloads.

Viewing Audit Logs

1. Go to Enterprise Admin → Audit Logs.
2. Filter by: date range, user, action type, or resource.
3. Click any entry to see full details including IP address and user agent.

Exporting Logs

Export audit logs in CSV or JSON format for external analysis or compliance reporting:

• Click "Export" in the top-right of the audit log view.
• Select format, date range, and filters.
• Logs can also be streamed to your SIEM via webhook.

Retention

Enterprise audit logs are retained for 1 year by default. Extended retention (up to 7 years) is available for compliance-regulated industries.