Skip to main content
Cover Image
Current Integration Issue: There are currently limitations with setting custom headers in the Gumloop MCP server configuration, which may prevent proper authentication with Membit’s API. We are working to resolve this issue.
Gumloop is an intuitive no-code platform that enables you to build AI-powered workflows and automations. By integrating Membit with Gumloop, you can enhance your AI agents with real-time social context, allowing them to access current trends, breaking news, and live conversations from across the web.

​
Prerequisites

Before you begin, make sure you have:
  • An active Gumloop account (sign up here)
  • A Membit account with an API key get your API key
  • Basic familiarity with Gumloop’s workflow builder

​
Setting Up Membit MCP Server

Follow these steps to integrate Membit with your Gumloop workflows:
1

Access Settings

Navigate to your Gumloop dashboard and click on the Settings option in the main menu.
Gumloop dashboard with settings menu highlighted
2

Go to Personal Credentials

In the settings menu, locate and click on Personal Credentials to manage your external service connections.
Gumloop settings page showing personal credentials section
3

Add New Credentials

Click the Add Credentials button to create a new credential set for external services.
Add credentials button in Gumloop personal credentials page
4

Search for MCP Server

In the credential type search, type “MCP server” and select the MCP Server option when it appears.
Searching for MCP server in Gumloop credentials options
The MCP Server credential type enables Gumloop to connect with external Model Context Protocol servers like Membit.
5

Configure MCP Server Settings

Fill in the MCP server configuration with the following details:
MCP server configuration form in Gumloop with URL and label fields
Replace <your-api-key> with your actual Membit API key. Keep this credential secure and don’t share it with unauthorized users.
Click Save to store your MCP server credentials.

​
Building Your First Workflow

Now that your Membit MCP server is configured, let’s create a workflow that uses real-time context:
1

Navigate to the Hub

Go to the Hub section in your Gumloop dashboard to access the workflow builder.
Gumloop hub interface showing available workflow options
2

Create New Flow

Click Create Flow to start building a new workflow from scratch.
3

Add Your First Node

Click Add your first node to begin building your workflow logic.
Gumloop workflow builder showing add first node option
4

Search for Ask AI

In the node search bar, type “Ask AI” to find the AI conversation node.
Searching for Ask AI node in Gumloop node library
5

Configure Ask AI Node

Select the Ask AI node and configure it with these settings:
Ask AI node configuration interface in Gumloop
Basic Configuration:
  • Prompt: “Tell me about crypto news using membit” (or your preferred query)
  • AI Model: Choose “Claude 4 Sonnet” or your preferred model
Ask AI node options showing MCP server connection settings
Advanced Options:
  1. Click “Show more options”
  2. Set “Connect MCP server” to Yes
  3. Choose MCP servers: Select the “membit-mcp” you created earlier
6

Test Your Workflow

Click Run to execute your workflow and see Membit’s real-time context in action.
Gumloop workflow execution results showing AI response with real-time context

Workflow execution showing real-time context from Membit

If successful, your AI agent will respond with current information about your query topic, powered by Membit’s real-time data feed.

​
Troubleshooting

Problem: Cannot connect to Membit MCP server in GumloopSolutions:Verify the MCP server URL is correct: https://mcp.membit.ai/mcpCheck your API key is correct and activeEnsure the Additional Header format is exact: X-Membit-Api-Key: <your-api-key>Try re-saving the MCP server credentialsCheck if Gumloop has proper internet access
Problem: API key authentication is failingSolutions:Verify your Membit API key is active in your dashboardEnsure there are no extra spaces or characters in the API keyCheck that the header name is exactly X-Membit-Api-Key (case-sensitive)Try generating a new API key if the current one isn’t workingConfirm your Membit account has sufficient credits/usage allowance
Problem: Ask AI node doesn’t use Membit toolsSolutions:Ensure “Connect MCP server” is set to YesVerify the “membit-mcp” server is selected in MCP servers dropdownTry switching to Claude 4 Sonnet model for better tool usageMake your prompt more explicit about using Membit: “Use membit to tell me about…”Check if the workflow has proper permissions to access external servers