Jina Web Search MCP

# ============================================================================= # Jina Web Search MCP Server - Docker Configuration # ============================================================================= # This Dockerfile creates a containerized MCP (Model Context Protocol) server # that provides web search and URL content fetching capabilities using Jina AI. # # Build: docker build -t jina-mcp-server . # Run: docker-compose up -d # ============================================================================= # Use Python 3.11 slim image for optimal size and security # Slim variant excludes unnecessary packages while maintaining compatibility FROM python:3.11-slim # Set working directory inside the container # All subsequent commands will run from this directory WORKDIR /app # Copy requirements first to leverage Docker's layer caching # This allows rebuilds to skip dependency installation if requirements haven't changed COPY requirements.txt . # Install Python dependencies # --no-cache-dir reduces image size by not storing pip cache RUN pip install --no-cache-dir -r requirements.txt # Copy application code after dependencies to optimize rebuild times # Changes to server.py won't trigger dependency reinstallation COPY server.py . # Expose the port that the MCP server runs on # This documents which port the application uses (required for docker-compose) EXPOSE 5003 # Set environment variables for optimal Python execution in containers ENV PYTHONUNBUFFERED=1 # PYTHONUNBUFFERED=1 ensures Python output is sent directly to terminal # This is crucial for proper logging in Docker containers # Note: JINA_API_KEY is intentionally not set here # It will be provided by docker-compose.yml with fallback handling: # ${JINA_API_KEY:-your_jina_api_key_here} # Run the server # Using exec form (list) for better signal handling and process management CMD ["python", "server.py"]