Overview

Firecrawl MCP is an MCP service that crawls any web page or entire website and delivers the content in an LLM-compatible, cleaned data format (Markdown, JSON, plain text, etc.).

This allows your Kuika application to pull content from the web in the form of structured data, which can be used directly in AI Agents, automation scenarios, and data processing workflows. With Firecrawl MCP integration:

• You eliminate the complexity of HTML on web pages,
• You obtain data outputs optimized for AI models,
• and scenarios such as content summarization, analysis, indexing, and chatbots are simplified.

Firecrawl MCP can only be used within AI Agents.

Defining Firecrawl MCP via Configuration Manager

Firecrawl MCP settings are defined via the Kuika Configuration Manager module.

Setup Steps

1. Log in to the Kuika platform.
2. Go to the Configuration Manager module from the left menu.
3. Open the Firecrawl MCP Setting field on the MCP Settings screen.
4. Fill in the following fields:

• Field Descriptions
  
  • Name (Required)
    
    • Enter a descriptive name for your Firecrawl configuration. E.g.:
    • Firecrawl_Main_Key
    • WebScraping_Service
    • Firecrawl_MCP_Prod
  • API Key (Required)
    
    • This field requires the API key obtained from the Firecrawl service.

Creating a Firecrawl API Key

You must create an API Key to use Firecrawl MCP.

Step-by-Step API Key Acquisition Process

1. Go to the following site in your browser: https://www.firecrawl.dev/

2. If you don't have an account yet, register with Sign Up.
3. Log in to your account after registering.
4. Select an appropriate Subscription Plan.
5. Go to the Dashboard screen.

6. Click on the API Keys tab in the right menu.

7. Find your generated API key on the screen that opens.
8. Click the Copy button to copy the key.

9. In your Kuika project:
   
   • Go to Configuration Manager → Firecrawl MCP Setting,
   • Paste the key you obtained into the API Key field.

Usage Scenarios

Firecrawl MCP is heavily used in the following scenarios:

• Analyzing website content with AI Agents,
• Scanning blogs, news, or documentation sites,
• Summarizing and classifying content,
• Creating search indexes,
• Providing live web data to chatbot systems,
• Creating automatic data update and scraping scenarios.

Using Firecrawl in AI Agent

Firecrawl MCP is used in the AI Agent screen:

• It is used in AI Agents under the name Dynamic Website Scraper tool.

Usage Flow

1. Go to the Datasources module.
2. Enter the AI Agent tab.
3. Add the Dynamic Website Scraper to the Agent using the Add Tool.
   
   • In the Tool settings: Select the Firecrawl MCP configuration you created earlier.
4. In the tool settings:
   
   • Select the Firecrawl MCP configuration you created earlier.
5. In Agent Instruction:
   
   • Write which sites will be crawled,
   • What type of data is expected,
   • How the data will be interpreted.
6. When the agent is run, the content is pulled from Firecrawl and included in the AI output.