Getting Started with Cypher LM

This path is for anyone new to Cypher LM — a founder, an office manager, a support lead — who wants the big picture before clicking around the dashboard. By the end of these five lessons you'll know what each product does, where to find it, and what to build first.

Cypher LM is a voice AI platform for businesses. Companies use it to build an AI agent that answers their customer calls instantly, 24 hours a day, in English or Urdu, grounded in the business's own information rather than generic chatbot filler. There is no call center behind it and no hold music in front of it — a caller dials in, or opens a website widget, and talks to an agent that already knows the business's services, pricing, and policies because you gave it that knowledge when you built it.

It is easy to assume a product like this is a calendar-booking tool, because scheduling is one of the things it can do. That is only one piece. The core of Cypher LM is the conversation itself — the agent understands what a caller wants and resolves it on the spot, whether that means answering a question, booking a slot, taking a payment, or simply logging the request for a human to follow up on.

Who builds an agent

• A clinic that wants every call answered on the first ring, even after hours.
• A recruiting agency that wants candidate calls screened and interview slots booked automatically.
• A small retailer that wants order status and return questions answered without a phone queue.
• A professional services firm that wants every call captured as notes and follow-ups, not lost in someone's memory.

None of these businesses need a developer. The agent is built through a guided, no-code wizard that takes about five to ten minutes, which the next path in this hub, Set Up Cypher LM Step by Step, walks through in detail. For now, keep reading — the next lesson introduces the five products by name so the rest of the dashboard makes sense the first time you see it.

It also helps to understand what makes the agent trustworthy enough to put in front of real customers. Every agent is grounded in the business's own material rather than guessing: when you build it, you describe your business, and you can upload real documents — a price list, a policy, an FAQ — that the agent reads from directly. If a caller asks something outside that material, the agent is designed to say so honestly rather than invent an answer, and to hand the conversation to a human when that's the better outcome. That combination — real knowledge plus honest limits — is what lets a business put an AI agent on its main phone line instead of treating it as a side experiment.

One more thing worth knowing before you go further: an agent, once built, is not a static recording. You can revisit its settings at any time from the Agents section of your dashboard, update what it knows, change its tone, or turn new features on as your needs grow. Getting started does not mean getting it perfect on the first try — it means getting something live, testing it yourself, and refining it from real conversations.

Cypher LM is one platform made of five named products, each handling a different part of a customer conversation. You don't have to use all five — a new account can start with just the voice agent and turn the others on later — but understanding what each one does will save you time in the dashboard.

Cypher V is the voice agent itself, and the flagship product. It is the part that actually picks up the call or the web chat, understands the caller, and answers from your business's own knowledge. Everything else in this list is something Cypher V can optionally do during that same call, once you turn the feature on for an agent.

• Cypher S — scheduling. Checks real availability and books an appointment into your calendar before the call ends, with a Google Meet link for video appointments and a confirmation sent automatically.
• Cypher M — the live copilot for your human team. Before you call someone back, it briefs you on everything they've said in past calls; during a live call, it listens and suggests what to say next.
• Cypher N — the note taker. Turns every call into a clean summary, action items, and a drafted follow-up email, so nothing said on a call gets lost.
• Cypher P — payments. Lets the agent request a payment mid-call, with a secure link the caller can pay through by card, bank transfer, JazzCash, or Easypaisa.

Two smaller pieces round out the platform. Cypher C keeps a contact record for every caller, pulling together their full call history so Cypher M and Cypher N have something to work from. Cypher Dots is a newer, beta feature that lets a customer text your agent over SMS instead of calling, for businesses that want that option.

Each product lives inside the same dashboard and the same agent — there's nothing extra to install. Turning one on is a toggle in the agent's settings, not a separate purchase or a new piece of software to learn, and turning it off again is just as easy if it turns out you don't need it yet.

It's worth noticing how these five pieces connect to each other, because that's where the real value shows up. A call answered by Cypher V can end with Cypher S booking a meeting, Cypher P collecting a deposit, and Cypher N writing the recap — all from one conversation, with no one typing anything into a spreadsheet afterward. Cypher M then uses that same history the next time someone on your team needs to call that person back. None of the products work in isolation; they all draw from the same record of what actually happened on the call. The next lesson walks through where to find each one once you're signed in.

Once you sign in at cypherlm.com/dashboard, everything you'll use day to day lives in one left-hand navigation. It can look busy the first time, so here is what each section is actually for.

• Home — a quick overview of recent activity across your agents.
• Agents — the list of agents you've built, and the settings for each one, including which features (booking, payments) are turned on.
• Bookings — every appointment booked on a call or through your public booking link, plus your availability and booking types.
• Calls — your full call log, with transcripts, recordings, and the Recaps tab for Cypher N summaries.
• Copilot — Cypher M: live sessions, playbooks, and a library of suggestions your team has used.
• Notes — the Cypher N recap and follow-up view, organized per call.
• Payments — the Cypher P overview, payment history, reusable payment links, and payment settings.
• Contacts — every caller, with their history across calls so you're never starting from zero.

Below those are Analytics, for trends across calls and bookings; Automations, where you switch on things like recap emails and booking reminders without touching anything manually; Deploy, for publishing an agent; Dots, for the SMS beta; and Settings, where Google and payment integrations, team roles, and your plan all live.

You don't need to visit every section before your first agent goes live. Agents and Calls are where you'll spend the most time early on — everything else becomes useful as your usage grows. The next lesson covers how an agent actually gets built, step by step, in overview.

A reasonable way to spend your first real session in the dashboard is to open Agents, look at the one agent you've built, and read through its settings end to end — the knowledge it has, the behaviour you defined, the features you've turned on. Then open Calls and watch what happens the next time someone actually talks to it. Seeing a real transcript next to the settings that produced it is the fastest way to understand how the pieces of the dashboard connect, faster than reading about any one section in isolation.

If you're setting this up for a small team rather than just yourself, it's worth looking at Settings early too, even before your first agent is fully tuned. That's where team roles live, and deciding early who can edit an agent's knowledge versus who can only view call history keeps things tidy as more people start relying on the account.

Building an agent happens through a single guided wizard, reachable from the "Build Your Agent" button on your dashboard or the homepage. It asks for information about your business in plain language — no configuration files, no code — and uses your answers to generate a complete agent.

At a high level, the wizard collects: your business details (name, website, what you do, who you serve), a contact number the agent can give out, the agent's identity (a name, a greeting, a tone of voice), how it should behave (what it should and shouldn't discuss, and what to say when it doesn't know an answer), and which language or languages it should speak. If you give it your website address, Cypher LM visits the site and pulls in content automatically, so you're reviewing pre-filled answers rather than starting from a blank page.

At the end of the wizard, Cypher LM generates a system prompt — the complete instruction set that defines how your agent talks and what it knows — and shows it to you before anything goes live. You can read it, edit it, or add knowledge files (a price list, an FAQ document, a policy PDF) so the agent answers from your real material instead of guessing.

Deployment itself is a single click. The agent is live in roughly ten to twenty seconds, and you can test it immediately from the dashboard by talking to it yourself before anyone else does. The full walkthrough, with every field explained, is in Set Up Cypher LM Step by Step — the next path in this hub.

The wizard is also where you choose language. Cypher LM supports English only, Urdu only, or both together in the same agent, switching naturally mid-conversation when a caller does. For a business serving a mixed audience, bilingual is usually the right default — it costs nothing extra to turn on and it means the agent never has to ask a caller to repeat themselves in a different language.

Because the whole process is guided, there's rarely a reason to over-plan before you start. Opening the wizard with a rough idea of your business and refining the details as you go produces a better result than waiting until every answer is perfect, since you can always come back and adjust the agent's knowledge, tone, or behaviour after it's already live and taking real calls.

Keep in mind that deploying is not the finish line — it's closer to opening day. The real tuning happens afterward, once real callers start asking questions you didn't anticipate while writing the wizard answers yourself.

Once an agent is live, it starts answering calls on its own, but a few things are worth understanding about what happens around each conversation.

Every call is logged automatically in Calls, including a transcript and, when enabled, a recording. If Cypher N is turned on for the agent, each call also gets a recap — a short summary, the key points raised, and any action items — within moments of the call ending. That recap can be emailed to you automatically, and you can send a drafted follow-up straight to the caller with one click instead of writing it from scratch.

If you've turned on booking or payments for an agent, those happen inside the call itself: a booking lands in Bookings with a calendar invite and Meet link already attached, and a payment shows up in Payments with its status (pending, paid, refunded) visible in real time. None of this requires you to do anything after the call — the automation toggles in Settings control which of these happen by default, so you can turn off anything you'd rather handle manually.

Analytics is the other place worth a regular look once an agent has handled a meaningful number of calls. It surfaces patterns across your whole call history — call volume over time, how often bookings and payments happen, where callers tend to ask the same question twice. Those patterns are the clearest signal for what to add to the agent's knowledge next: if the same question keeps coming up and the agent keeps falling back to its generic answer, that's a sign the underlying document is missing something, and uploading an updated file closes the gap immediately.

Treat the first week after deployment as a listening period rather than a finished product. The agent will do most things right immediately, but the small misses — a slightly wrong phone number, a tone that doesn't quite match your brand — only show up once real customers start talking to it, and they're quick to fix once you notice them.

Where to go next

If you haven't built an agent yet, Set Up Cypher LM Step by Step walks through the wizard field by field. If you want the agent to book meetings during the call, see Let Cypher S Book and Confirm Meetings for You. And if your team handles a high volume of calls, Make Every Meeting Count covers how Cypher M and Cypher N work together before, during, and after a conversation.