# Landbot MCP

> Landbot MCP lets your AI client manage entire conversational funnels from start to finish. You can pull detailed customer records, view full chat histories, send automated follow-up messages, or instantly hand off a conversation from a bot to a live agent—all without leaving your current workspace. It gives your agents the ability to act directly within complex customer support pipelines.

## Overview
- **Category:** communication-messaging
- **Price:** Free
- **Tags:** chatbots, conversational-ai, lead-capture, automated-messaging, customer-engagement, no-code

## Description

Managing customer conversations used to mean jumping between five different dashboards just to get context: one for chat history, one for lead status, another for active bots, and maybe a fourth just to send a quick follow-up text. This MCP changes that. It connects your AI agent directly into the Landbot system, letting it see and manipulate live conversations as if it were a human operator sitting at the desk.

Your agent can search for specific customers by email or pull up an entire chat sequence log to understand exactly what happened before you got involved. If the conversation is complicated, your agent can programmatically reassign that thread directly to a specialized human agent using simple commands. Plus, it lets you send automated text messages right into the active dialogue channel. Because this MCP lives in the Vinkius catalog, your AI client connects once and gets access to all these critical communication tools for your whole operation.

## Tools

### assign_agent
Programmatically hands off a conversation thread from the bot to a live human agent.

### get_bot
Retrieves specific details for one Landbot instance using its unique identifier.

### get_customer
Fetches the core profile and metadata for a single customer account.

### get_messages
Pulls the complete, chronological transcript of all messages exchanged with a customer.

### list_bots
Lists every Landbot active in your organization's account.

### list_customers
Retrieves a list of the most recently interacting customers with the bots.

### search_customers
Searches for and locates a specific customer profile using their email address.

### send_text_message
Sends an automated message directly into the ongoing conversation with a customer.

## Prompt Examples

**Prompt:** 
```
List standard bots running active pipelines right now.
```

**Response:** 
```
Loading Landbot configurations. There are currently 3 standard matrix bots running active logic mapping sequences.
```

**Prompt:** 
```
Fetch the entire transcription log for customer ID 98453.
```

**Response:** 
```
Extracted session log for 98453. The prospect answered 4 initial screening questions mentioning 'Price points'.
```

**Prompt:** 
```
Force assign the highest severity angry customer ticket to Agent Sarah.
```

**Response:** 
```
Operation complete. I've intercepted the live interaction and injected the routing update pointing to Agent Sarah gracefully.
```

## Capabilities

### List available bots
View a list of every bot operating within the Landbot system.

### Get specific bot details
Fetch detailed information about one particular bot by its unique ID.

### Find customer records
Search for a specific customer profile using an email address.

### Retrieve customer data
Pull core details and metadata for a single identified customer.

### Fetch chat message history
Download the complete chronological transcript of messages exchanged with a customer.

### Hand off conversation to staff
Programmatically reroute an active bot dialogue and assign it to a live human agent.

### Send automated text replies
Send a structured message directly into the customer's ongoing chat thread.

## Use Cases

### A sales rep needs to qualify a new lead.
The rep doesn't want to rely on memory. They prompt their agent to search for the prospect by email using `search_customers`. The agent instantly pulls up the contact’s full profile, allowing the rep to immediately reference specific details—like which bot they interacted with last week—and ask highly targeted questions.

### Support needs to escalate a complex technical issue.
The initial chatbot interaction failed because it hit a knowledge gap. The support agent tells their AI client to run `get_messages` for the customer ID, pulling up every single message exchanged. This history allows the agent to understand the full scope of failure and then use `assign_agent` to hand off the case with a perfect summary.

### Product team needs to analyze chat bottlenecks.
The product manager wants to know why customers drop off. They run `list_bots` first, identifying all active funnels. Then they use `get_bot` on a specific bot and cross-reference that data with customer profiles (`list_customers`) to pinpoint exactly where the conversation fails or gets stalled.

### Marketing needs to re-engage an inactive lead.
The marketing team identifies high-value leads who went silent. They use `get_customer` to pull up their last known details, then instruct the agent to send a personalized message using `send_text_message`, restarting the conversation loop without human intervention.

## Benefits

- Instantly see full context. Instead of asking a customer, 'What were we talking about?', your agent runs `get_messages` and has the entire chat transcript ready to analyze and respond to, ensuring zero friction in service.
- Control the handoff process. If the AI can't solve it, you don't want the conversation to stall. Use `assign_agent` to immediately route a complex ticket to a human agent, guaranteeing continuity of care.
- Deep customer profiling. Need more than just a name? Running `get_customer` retrieves key metadata about the account, letting your agent tailor responses based on subscription level or purchase history.
- Targeted outreach. You can't wait for a lead to come back online. Use `send_text_message` to send a specific follow-up message directly into their chat channel when they need it most.
- Operational oversight. By using `list_bots`, your team can quickly audit which conversational pathways are running and check the status of every bot without logging into separate monitoring dashboards.

## How It Works

The bottom line is: you use natural language prompts in your AI client instead of logging into multiple service dashboards.

1. First, you authorize your AI client connection through Vinkius and provide the Landbot API token in the credentials.
2. Next, you instruct your agent to perform a specific action, like fetching chat logs or rerouting an interaction. The MCP executes that command against the Landbot system.
3. Finally, your agent receives structured data—whether it's the customer's full message history, their profile details, or confirmation that the conversation has been assigned to Sarah.

## Frequently Asked Questions

**How do I use Landbot MCP to see all my active bots?**
You run `list_bots` to get an inventory of every bot running in your account. This is the first step if you need to audit or understand what conversational pipelines exist.

**What information does Landbot MCP provide about a customer?**
You can use `get_customer` to pull core profile details and metadata on a single individual. You also run `search_customers` if you only know their email address.

**Can I send messages through the Landbot MCP?**
Yes, you can use `send_text_message` to programmatically inject an automated reply into a customer's live chat conversation. This is useful for follow-ups or confirming information.

**How do I make sure a difficult chat gets to a human?**
Use the `assign_agent` tool. Your agent intercepts the dialogue and routes it away from bot control, assigning ownership to a live team member immediately.

**Is Landbot MCP only for viewing data?**
No, it's much more than that. You can perform actions like sending messages and reassigning agents, making it an active control layer over your customer service system.