Everything you need to know about ModusOp AI Chat.
ModusOp AI Chat is an AI-powered chat widget and voice receptionist you can add to your website. It answers customer questions using your actual website content, 24/7 — no manual training required.
The AI reads your llms.txt file (a simple text file at your site root) and crawls your website pages to build a knowledge base. Content is re-indexed daily so the AI always has current information.
llms.txt is a simple text file you place at the root of your website (like robots.txt). It provides structured information about your business specifically for AI systems. Having one gives the AI better context and more accurate answers than crawling alone. Learn more.
Not at all. After signing up, you add one line of JavaScript to your website. Everything else — content indexing, AI responses, conversation management — is handled automatically and configured through a simple admin dashboard.
Yes. From your admin dashboard you can customise primary colour, secondary colour, background colour, widget position (bottom-left or bottom-right), welcome message, placeholder text, and the AI's system prompt to match your brand and tone.
The voice receptionist answers phone calls using AI with natural, sub-500ms response times. It uses the same knowledge base as the chat widget to answer caller questions. It can transfer calls to your team during business hours and take voicemails when you're unavailable.
Outside your configured business hours, the voice AI can still answer general questions. If the caller needs to speak with someone, it offers to take a voicemail. The recording and transcription are emailed to you and available in your admin dashboard.
The AI is designed to be honest. If it doesn't have enough information to answer a question, it tells the visitor and directs them to your contact page so a real person can help. It never makes up or fabricates information.
Yes. Enterprise plan users can configure their own OpenAI API key in the admin dashboard for full control over API usage and billing. Otherwise, a shared key is used and included in your plan.
You can choose from OpenAI models including GPT-4, GPT-4o, and GPT-4o-mini. The model can be configured per-tenant from your admin dashboard. Voice calls default to GPT-4o-mini for faster response times.
Yes. All data is encrypted in transit via HTTPS, API keys are stored securely on the server, and conversations are only accessible to the tenant who owns them. We do not share your data with third parties. Admin accounts support two-factor authentication (TOTP).
Billing is handled through Stripe with monthly subscriptions. You can manage your subscription, update payment details, and view invoices directly from your admin dashboard or Stripe's customer portal.