Skip to main content
Glama

Model Control Plane (MCP) Server

services: # This is an alternative docker-compose configuration that uses bridge networking # instead of host networking mode. Use this if you need to communicate with # other Docker services but also reach the host's MCP server. langflow: image: langflowai/langflow:latest container_name: langflow # Use bridge networking (default) # network_mode: bridge # This is the default so we don't need to specify it ports: # Standard langflow UI port - "7860:7860" # Forward host's MCP port to the same port inside the container # This allows container to access host's MCP server via localhost:8000 - "127.0.0.1:8000:8000" volumes: - ./components:/home/user/app/src/components environment: - LANGFLOW_COMPONENTS_PATH=/home/user/app/src/components/custom/tools - LANGFLOW_LOG_LEVEL=DEBUG - LANGFLOW_AUTO_RELOAD=true - DOCKER_HOST_IP=172.17.0.1 extra_hosts: - "host.docker.internal:host-gateway" command: langflow run --host 0.0.0.0 --port 7860 --backend-only --dev restart: unless-stopped networks: - monitoring networks: monitoring: external: true

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/dvladimirov/MCP'

If you have feedback or need assistance with the MCP directory API, please join our Discord server