telecrm AI

Before you start generating the telecrm reports via MCP, let's make sure you have everything in place. In this video, we will cover the prerequisites needed to enable MCP, and we'll walk through the configuration — right up to managing your MCP connection. No coding, no API keys, just a few minutes.

Let's get started.

Section 1 — Prerequisites

Before you start, make sure the following are in place:

• Workspace admin access to telecrm. You'll need admin rights to open the MCP Setup page and generate a token. From there, you will have to request access to get started, which is covered in the next section.
• An MCP-aware AI client. This is the assistant you'll connect telecrm to. Right now, that's either Claude or ChatGPT. A few things to keep in mind when choosing:
• You can get started with the free version of Claude.
• ChatGPT requires a paid (Plus, Teams, or Enterprise) plan to use MCP.
• For the best experience, the Pro version of Claude is recommended. On the free version, prompts expire quickly, so complex queries may not generate a full, complete output.

That's all you need to get started. There are no API keys to hunt down and no tokens to create manually — the MCP Integrations page takes care of everything from here.

Section 2 — Open MCP Settings and Request Access

1. In telecrm web app, go to the side navigation menu bar and click on the MCP icon.
2. You can now see the MCP page. Click on the Request access option. A dialog opens with a pre-filled email to your account manager, with your workspace name and ID included automatically.
3. Click Open in mail app to send it through your default mail client, or use Copy full email to send it manually. Approval typically takes up to 2 business days max. 
4. Once approved, the page changes to show an Enable MCP button.

Section 3 — Enable MCP and  Pick Your AI Provider

1. On the MCP page, you can watch the 2-minute walkthrough video first if you'd like a visual overview.
2. Click Enable MCP
3. Click the tile for the AI you'll connect first: Claude or ChatGPT. You can change this later and connect multiple providers by creating extra tokens after setup.
4. Click Continue

Section 4 — Create Your Access Token

1. A default token name is generated for you (it looks like yourworkspace_provider_yourname_read_…). 
2. Click Generate. The next screen shows your token once. Use Copy to put it on your clipboard, or Download to save it as a .txt File named after your workspace.
3. Treat this token like a password. telecrm only stores a hashed version and cannot recover the plaintext if you lose it — you'd have to create a new one and reconnect. Save it in your password manager or a secure note before clicking Continue.

Section 5 — Add Workspace Context

1. This is where your AI assistant gets to know your business. This step is crucially important and must not be missed since it sets a focus and direction as to how the AI will read your data. The sections are:

• A — About company: who you are, what you sell, who your customers are. 
• B — Business process: what your team does (sales, support, collections)
• S — Source of lead: where your leads come from or the main Lead origins.
• J — Sales journey: your stages from first contact to close. This must include the entire customer journey from the moment of lead capture to its closure.
• I — Important notes: any hard rules (for example, "Never share lead phone numbers in answers or hide sensitive data like emails or phone numbers of leads.").

1. Click on Continue

And that's all you need to understand about the prerequisites and the initial activation process of MCP. From here, you can open Claude or ChatGPT, which will be covered in the next videos.

A few things to remember: keep your access token safe like a password, and add more tokens anytime from the MCP Settings page. The more workspace context you add, the sharper your assistant's answers will be — so it's worth revisiting that section as your business evolves.