Integrate with AI

Every page of this documentation is also available to AI coding assistants through our Model Context Protocol (MCP) server. Point your editor at it once, then ask your assistant to “integrate with Flabs” — it reads the live API reference, auth flow, schemas and webhook specs on demand and writes the integration code for you.

MCP server URL: https://developer.flabslis.com/mcp

How it works

The MCP exposes the same content you’re reading now — the auth token flow, CRM lead/booking endpoints, the billing pre-booking API, webhook events and the OpenAPI specs — as structured, searchable context. Your assistant pulls only the pages it needs, so it generates code against the current API instead of guessing.

1. Add the MCP server to your editor

Pick your tool. The server is a remote HTTP MCP — no install, no API key needed to read the docs.

Cursor
Claude Code
VS Code (Copilot)
Windsurf

Add to .cursor/mcp.json in your project (or the global config):

{
  "mcpServers": {
    "flabs-developer": {
      "url": "https://developer.flabslis.com/mcp"
    }
  }
}

Run in your project directory:

claude mcp add --transport http flabs-developer https://developer.flabslis.com/mcp

Add to .vscode/mcp.json:

{
  "servers": {
    "flabs-developer": {
      "type": "http",
      "url": "https://developer.flabslis.com/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "flabs-developer": {
      "serverUrl": "https://developer.flabslis.com/mcp"
    }
  }
}

Reload your editor so it picks up the new server. You should see a flabs-developer MCP with two tools: a docs search and a read-only docs filesystem.

2. Ask the AI to integrate

Once the server is connected, describe what you want in plain language. The assistant will search the docs, read the relevant endpoints, and scaffold the code. A few prompts that work well:

Using the flabs-developer MCP, add a Flabs integration to this project:
authenticate with my client credentials, fetch my branches, and create a CRM lead.

Set up a webhook handler for Flabs bill.completed and report.completed events,
following the payload structure in the flabs-developer docs.

I'm a corporate (party-scoped) account. Wire up test-pricing lookups and lead
creation scoped to my corporateUser, per the Flabs party-scoped guide.

The assistant pulls the exact request/response shapes, required fields, rate limits and error codes from the docs — so the generated client matches the real API.

The MCP only exposes public documentation. Never put your clientID, clientSecret, or tokens into prompts or commit them to source control — load them from environment variables or a secrets manager. See the Authentication guide.

What the assistant can build

Authentication

Token generation and refresh handling with your client credentials.

CRM leads & bookings

Branch lookup, test/package pricing, and lead/booking creation.

Billing pre-bookings

Create pre-bookings that kick off the lab workflow.

Webhook handlers

Endpoints that react to bill, report and CRM events.

Tips for better results

Tell it your account mode. Mention whether you’re catch-all (default) or party-scoped so it scopes calls correctly.
Name your stack. “in Node/Express”, “in Python/FastAPI”, etc. produces idiomatic code.
Ask it to verify against the docs. “Double-check the required fields against the flabs-developer MCP” catches missed parameters before you run anything.

Support

Stuck or need credentials? Reach our team at hello@flabs.in or +91 7253928905, Monday to Friday, 9 AM to 6 PM IST.

Flabs Developer API guide

Authorization

Webhooks

Billing

Patient Reports

CRM

How it works

1. Add the MCP server to your editor

2. Ask the AI to integrate

What the assistant can build

Authentication

CRM leads & bookings

Billing pre-bookings

Webhook handlers

Tips for better results

Support

​How it works

​1. Add the MCP server to your editor

​2. Ask the AI to integrate

​What the assistant can build

Authentication

CRM leads & bookings

Billing pre-bookings

Webhook handlers

​Tips for better results

​Support

How it works

1. Add the MCP server to your editor

2. Ask the AI to integrate

What the assistant can build

Tips for better results

Support