MCP Chatbot
About
An intelligent chatbot for automating tasks like browser control, web searches, and travel planning.
Details
- Author
- naakaarafr
- Categories
- Productivity, Other, AI
Jump to
Setup
Install MCP Chatbot in your MCP client (Claude Desktop, Cursor, Windsurf, and others).
Repository: https://github.com/naakaarafr/MCP-Chatbot
Follow the installation instructions in the repository README, then restart your MCP client.
MCP Chatbot - Intelligent automation through Model Context Protocol integration. Features browser automation, web search, and travel planning with robust error handling. Uses Google Gemini LLM and gracefully handles server failures for reliable operation.
GitHub Repository:https://github.com/naakaarafr/MCP-Chatbot
A powerful chatbot built with Model Context Protocol (MCP) that provides intelligent automation through browser control, web search, and travel planning capabilities.
This chatbot leverages the Model Context Protocol to integrate multiple specialized servers, creating a unified interface for web automation, search, and travel-related tasks. It uses Google's Gemini model as the underlying LLM and can gracefully handle server failures by testing each MCP server individually.
- Navigate to websites
- Take screenshots
- Click elements and interact with web pages
- Automate web-based tasks
- Perform web searches
- Get real-time information
- Research topics and find answers
- Search for accommodations
- Find hotels and rentals
- Get travel recommendations
- Individual server testing on startup
- Graceful degradation when servers fail
- Fallback responses in limited mode
- Clear error reporting
- Python 3.8+
- Node.js and npm (for MCP servers)
- UV package manager (for Python MCP servers)
# Python dependencies pip install asyncio python-dotenv langchain-google-genai mcp-use # Node.js MCP servers npm install -g @playwright/mcp npm install -g @openbnb/mcp-server-airbnb # Python MCP servers uvx install duckduckgo-mcp-server
# Ensure you have both app.py and browser_mcp.json
pip install asyncio python-dotenv langchain-google-genai mcp-use
# Playwright for browser automation npm install -g @playwright/mcp # Airbnb for travel search npm install -g @openbnb/mcp-server-airbnb # DuckDuckGo for web search uvx install duckduckgo-mcp-server
Set up environment variables:Create a.envfile in the project root:
Configure MCP servers:Thebrowser_mcp.jsonfile is already configured with default settings. Modify if needed:
{ "mcpServers": { "playwright": { "command": "npx", "args": ["@playwright/mcp@latest"] }, "airbnb": { "command": "npx", "args": ["-y", "@openbnb/mcp-server-airbnb"] }, "ddg-search": { "command": "uvx", "args": ["duckduckgo-mcp-server"] } } }
- help- Show available commands and example queries
- tools- List all available MCP tools
- quitorexit- End the conversation
"Take a screenshot of google.com" "Navigate to wikipedia.org and search for Python" "Click the login button on the website"
"What's the weather like today?" "Search for the latest news about AI" "Find information about renewable energy"
"Find hotels in New York City" "Search for Airbnb rentals in San Francisco" "What are the best accommodations in Tokyo?"
"Find the best restaurants in downtown Chicago" "Help me research Python programming tutorials" "Take a screenshot of the current stock market trends"
-
MCPChatbot Class: Main chatbot orchestrator
- Initializes and manages MCP servers
- Handles user interactions
- Provides fallback responses
Server Testing: Individual server validation
- Tests each MCP server on startup
- Reports working and failed servers
- Continues operation with available servers
Agent Integration: LangChain + MCP integration
- Uses Google Gemini 2.0 Flash as the LLM
- Integrates with MCP client for tool access
- Supports up to 20 reasoning steps
The chatbot implements robust error handling:
- Server Failures: Individual server testing prevents total failure
- Limited Mode: Provides basic responses when no servers are available
- Graceful Degradation: Continues operation with partial functionality
- User Feedback: Clear error messages and status reporting
Editbrowser_mcp.jsonto add or modify MCP servers:
{ "mcpServers": { "your-server-name": { "command": "command-to-run", "args": ["arg1", "arg2"] } } }
- GOOGLE_API_KEY: Your Google AI API key for Gemini access
self.llm = ChatGoogleGenerativeAI( model="gemini-2.0-flash-exp", # Model selection temperature=0.3 # Creativity level )
- Ensure the configuration file is in the same directory asapp.py
- Check that all MCP servers are properly installed
- Verify Node.js and UV are installed and accessible
- Check server logs for specific error messages
- Ensure your.envfile contains a validGOOGLE_API_KEY
- Verify the API key has access to Gemini models
- Make sure Node.js and npm are installed
- Install UV package manager:pip install uv
- Try installing MCP servers individually to isolate issues
The chatbot provides detailed startup information:
- Server testing results
- Working server list
- Error messages for failed servers
- Fork the repository
- Create a feature branch
- Make your changes
- Test with various MCP servers
- Submit a pull request
This project is licensed under the MIT License - see theLICENSEfile for details.
Please check individual MCP server licenses for their respective terms.
- MCP Protocol: Check the official MCP documentation
- Specific MCP Servers: Refer to individual server documentation
- LangChain Integration: See LangChain documentation
- Google Gemini: Check Google AI documentation
- Model Context Protocol (MCP) team for the innovative protocol
- Playwright team for browser automation capabilities
- LangChain team for the integration framework
- Google for providing Gemini AI models
Business management platform controlled entirely through conversation. Describe your business and AI builds your complete system — tables, views, dashboards, chatbot — in 2 minutes. 40+ MCP tools, 26 industry templates, free. Remote server with OAuth 2.1 + PKCE at https://runikapp.com/mcp
A cross-platform AI chat client supporting desktop, mobile, and web platforms.
Interact with task, doc, and project data in Dart, an AI-native project management tool
Create notes, search, & think with your Fabric AI workspace
Great Question is an Agentic UX research platform for product builders. Its MCP lets AI agents create studies directly from any AI tool, surface insights, find the right research candidates, and query your entire research repository.
Magica is your all-in-one AI platform, offering 2500+ cutting-edge tools under a single subscription.
Prompt Buddy MCP exposes a public, searchable catalog of reusable AI skills.
Connect to the Taskade platform via MCP. Access tasks, projects, workflows, and AI agents in real-time through a unified workspace and API.
Dynamically search and call tools using UnifAI Network
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
