GPT Researcher

While LLM apps can access web search tools with MCP,GPT Researcher MCP delivers deep research results.Standard search tools return raw results requiring manual filtering, often containing irrelevant sources and wasting context window space.

GPT Researcher autonomously explores and validates numerous sources, focusing only on relevant, trusted and up-to-date information. Though slightly slower than standard search (~30 seconds wait), it delivers:

- ✨ Higher quality information
- 📊 Optimized context usage
- 🔎 Comprehensive results
- 🧠 Better reasoning for LLMs

https://github.com/user-attachments/assets/ef97eea5-a409-42b9-8f6d-b82ab16c52a8

git clone https://github.com/assafelovic/gptr-mcp.git pip install -r requirements.txt

{ "mcpServers": { "gptr-mcp": { "command": "python", "args": ["/absolute/path/to/gpt-researcher/gptr-mcp/server.py"], "env": { "OPENAI_API_KEY": "your-openai-key-here", "TAVILY_API_KEY": "your-tavily-key-here" } } } }

Restart Claude Desktopand start researching! 🎉

For detailed setup instructions, see thefull Claude Desktop Integration sectionbelow.

- deep_research: Performs deep web research on a topic, finding the most reliable and relevant information
- quick_search: Performs a fast web search optimized for speed over quality, returning search results with snippets. Supports any GPTR supported web retriever such as Tavily, Bing, Google, etc... Learn morehere
- write_report: Generate a report based on research results
- get_research_sources: Get the sources used in the research
- get_research_context: Get the full context of the research

You can also connect any other web search engines or MCP using GPTR supported retrievers. Check out thedocs here

git clone https://github.com/assafelovic/gpt-researcher.git cd gpt-researcher

cd gptr-mcp pip install -r requirements.txt

OPENAI_API_KEY=your_openai_api_key TAVILY_API_KEY=your_tavily_api_key

# Build and run with docker-compose docker-compose up -d # Or manually: docker build -t gptr-mcp . docker run -d \ --name gptr-mcp \ -p 8000:8000 \ --env-file .env \ gptr-mcp

# First, start the container docker-compose up -d # Then connect to your n8n network docker network connect n8n-mcp-net gptr-mcp # Or create a shared network first docker network create n8n-mcp-net docker network connect n8n-mcp-net gptr-mcp

Once the server is running, you'll see output indicating that the server is ready to accept connections. You can verify it's working by:
- SSE Endpoint: Access the Server-Sent Events endpoint athttp://localhost:8000/sseto get a session ID
- MCP Communication: Use the session ID to send MCP messages tohttp://localhost:8000/messages/?session_id=YOUR_SESSION_ID
- Testing: Run the test script withpython test_mcp_server.py

# Local development (default) python server.py # ➜ Uses STDIO transport (Claude Desktop compatible) # Docker environment docker run gptr-mcp # ➜ Auto-detects Docker, uses SSE transport # Manual override export MCP_TRANSPORT=sse python server.py # ➜ Forces SSE transport

// ~/Library/Application Support/Claude/claude_desktop_config.json { "mcpServers": { "gpt-researcher": { "command": "python", "args": ["/absolute/path/to/server.py"], "env": { "..." } } } }

# Set transport explicitly for web deployment export MCP_TRANSPORT=sse python server.py # Or use Docker (auto-detects) docker-compose up -d

# Use the container name as hostname docker run --name gptr-mcp -p 8000:8000 gptr-mcp # In n8n, connect to: http://gptr-mcp:8000/sse

- Health Check:GET /health
- SSE Endpoint:GET /sse(get session ID)
- MCP Messages:POST /messages/?session_id=YOUR_SESSION_ID
- Local Development: Use default STDIO for Claude Desktop
- Production: Use Docker with automatic SSE detection
- Testing: Use health endpoints to verify connectivity
- n8n Integration: Always use container networking with Docker
- Web Deployment: Consider Streamable HTTP for modern clients

You can integrate your MCP server with Claude using:

Claude Desktop Integration- For using with Claude desktop application on Mac

{ "mcpServers": { "gptr-mcp": { "command": "python", "args": ["/absolute/path/to/your/server.py"], "env": { "OPENAI_API_KEY": "your-actual-openai-key-here", "TAVILY_API_KEY": "your-actual-tavily-key-here" } } } }

🔒 Your Claude Desktop config contains sensitive API keys.Protect it:

chmod 600 ~/Library/Application\ Support/Claude/claude_desktop_config.json

Never commit this file to version control.

Alternative: Environment Variable Script

For better security, create a wrapper script:

#!/bin/bash source /path/to/your/.env python /absolute/path/to/server.py

{ "mcpServers": { "gptr-mcp": { "command": "/absolute/path/to/run_gptr_mcp.sh" } } }

For complete step-by-step instructions, see theClaude Desktop Integration guide.

User: I'm considering investing in NVIDIA. Can you research the current state of the company, recent stock performance, latest product announcements, and analyst opinions? Claude: I'll help you research NVIDIA for investment purposes. This will take a moment as I conduct deep research to find the most up-to-date information about their stock performance, recent developments, and expert analysis. [Claude uses deep_research tool - this takes 30-40 seconds] Claude: Based on my research, here's a comprehensive analysis of NVIDIA's current state: ## NVIDIA (NVDA) Current Status ### Recent Stock Performance - Current price: $942.89 (as of my research) - YTD performance: +90.4% - 52-week range: $460.30 - $974.00 - Market cap: ~$2.32 trillion ### Recent Key Developments 1. Blackwell AI Platform: At GTC 2024, NVIDIA unveiled the Blackwell architecture, their next-generation AI chip platform promising 4x performance improvement over Hopper. Major cloud providers (AWS, Google, Microsoft) have already committed to integrating Blackwell. 2. Q1 FY2025 Earnings (May 2024): - Revenue: $26.0 billion (+262% YoY) - Data Center revenue: $22.6 billion (+427% YoY) - Gross margin: 78.4% - Announced 10-for-1 stock split effective June 7, 2024 3. Supply Chain Expansion: Increased investments in manufacturing capacity with TSMC to address ongoing chip shortages for their high-demand AI GPUs. ### Analyst Consensus - Strong Buy consensus with 37 analysts rating it "Buy," 5 "Hold," and 0 "Sell" - Average price target: $1,042 (post-split adjusted: $104.20) - Key bull thesis: Continued AI infrastructure spending and NVIDIA's dominant position in AI chips - Key bear concerns: Valuation, potential competition from AMD, Intel, and custom silicon from tech giants ### Industry Position - Dominant in AI GPU market with estimated 80-90% market share - Expanding ecosystem through CUDA software platform and industry partnerships - Diversifying beyond chips with DGX Cloud, AI software platforms, and automotive solutions

If you encounter issues while running the MCP server:
- API Keys: Make sure your API keys are correctly set in the.envfile
- Python Version: Check that you're using Python 3.11 or higher (required by gpt-researcher >=0.14.0)
- Dependencies: Ensure all dependencies are installed correctly:pip install -r requirements.txt
- Server Logs: Check the server logs for error messages

- Verify the container is running:docker ps | grep gptr-mcp
- Check container logs:docker logs gptr-mcp
- Confirm the server is binding to 0.0.0.0:8000 (logs should show this)

- Ensure both containers are on the same Docker network
- Use the container namegptr-mcpas the hostname in n8n
- Set the MCP server URL to:http://gptr-mcp:8000/sse

- The server uses SSE transport which requires session management
- First, get a session ID by connecting to/sseendpoint
- Use the session ID in subsequent MCP requests:/messages/?session_id=YOUR_ID
- Each client needs its own session ID

curl http://gptr-mcp:8000/sse # Look for: data: /messages/?session_id=XXXXX

curl -X POST http://gptr-mcp:8000/messages/?session_id=YOUR_SESSION_ID \ -H "Content-Type: application/json" \ -d '{"jsonrpc": "2.0", "id": 1, "method": "initialize", "params": {"protocolVersion": "2024-11-05", "capabilities": {"roots": {"listChanged": true}}, "clientInfo": {"name": "n8n-client", "version": "1.0.0"}}}'

curl -X POST http://gptr-mcp:8000/messages/?session_id=YOUR_SESSION_ID \ -H "Content-Type: application/json" \ -d '{"jsonrpc": "2.0", "id": 2, "method": "tools/call", "params": {"name": "quick_search", "arguments": {"query": "test"}}}'

Run the included test script to verify functionality:

- SSE connection and session ID retrieval
- MCP initialization
- Tool discovery and execution

If your MCP server isn't working with Claude Desktop:

- Check yourclaude_desktop_config.jsonsyntax is valid JSON
- Ensure you're usingabsolute paths(not relative)
- Verify the path toserver.pyis correct
- Restart Claude Desktop completely

"OPENAI_API_KEY not found" error:

- Make sure you added API keys to theenvsection in your config
- Don't forgetbothOPENAI_API_KEYandTAVILY_API_KEY
- API keys should be the actual keys, not placeholders

- Look for the 🔧 tools icon in Claude Desktop
- Check that Claude Desktop config file is in the right location:

- macOS:~/Library/Application Support/Claude/claude_desktop_config.json
- Windows:%APPDATA%\Claude\claude_desktop_config.json

- Make sure Python is accessible from the command line:python --version
- Try using full Python path:"command": "/usr/bin/python3"or"command": "python3"
- Check file permissions on your server.py file

- Test the server manually:python server.py(should show STDIO transport message)
- Check Claude Desktop logs (if available)
- Try the alternative script method from the integration section above

- Explore theMCP protocol documentationto better understand how to integrate with Claude
- Learn aboutGPT Researcher's core featuresto enhance your research capabilities
- Check out theAdvanced Usageguide for more configuration options

- Community Discord
- Email:assaf.elovic@gmail.com

Lightning-Fast, High-Accuracy Deep Research Agent 👉 8–10x faster 👉 Greater depth & accuracy 👉 Unlimited parallel runs