Get the FREE Ultimate OpenClaw Setup Guide →

tanzu

Tanzu MCP Server - AI-powered Cloud Foundry operations via Model Context Protocol

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio 0pens0-tanzu-mcp-server docker run -i tanzu-mcp-server:0.1.0 \
  --env CF_ORG="Your CF organization (optional)" \
  --env CF_SPACE="Your CF space (optional)" \
  --env CF_APIHOST="Cloud Foundry API host (e.g., api.your-cf-domain.com)" \
  --env CF_PASSWORD="Your CF password" \
  --env CF_USERNAME="Your CF username"

How to use

This MCP server provides access to Cloud Foundry operations through 38 AI-powered tools. It is built with Spring AI and exposes a streamable HTTP MCP interface, enabling clients to query, manage, and orchestrate CF resources via natural language prompts. After deploying, configure the server URL in your MCP client configuration. The tools are organized into categories such as Application Management, Organization & Space Management, Service Management, Route Management, Network Policy Management, Application Cloning, Target Management, and Configuration & Validation. You can ask the assistant to list apps, deploy or scale applications, manage spaces and routes, handle services, or validate CF configuration, all while benefiting from automatic retries, health monitoring, and token management baked into the server.

How to install

Prerequisites:

  • Docker installed and running
  • Access to a Docker registry containing the tanzu-mcp-server image (or build the image locally)

Installation steps:

  1. Pull the Docker image (or prepare for build): docker pull tanzu-mcp-server:0.1.0

  2. Run the server in Docker with required Cloud Foundry credentials (environment variables can be adjusted later): docker run -d
    --name tanzu-mcp-server
    -e CF_APIHOST=api.your-cf-domain.com
    -e CF_USERNAME=your-username
    -e CF_PASSWORD=your-password
    -e CF_ORG=your-organization
    -e CF_SPACE=your-space
    -p 8080:8080
    tanzu-mcp-server:0.1.0

  3. Verify the server is running: docker ps | grep tanzu-mcp-server curl -sS http://localhost:8080/actuator/health

  4. If you prefer local development, you can build and run with Maven (Java-based, not required if using Docker): ./mvnw clean package -DskipTests java -jar target/tanzu-mcp-server-0.1.0.jar

  5. Optional: configure a manifest-based deployment for Cloud Foundry or other platforms as appropriate to your environment.

Additional notes

  • Environment variables govern Cloud Foundry connectivity and are validated on startup. Missing required credentials (CF_APIHOST, CF_USERNAME, CF_PASSWORD) will fail fast with a clear error message.
  • The server uses HTTP Streamable transport and exposes a health endpoint at /actuator/health for observability.
  • The MCP client configuration should point to the deployed server URL, including the streamable transport method and appropriate timeout settings.
  • If you need to update CF target (organization/space), you can use the targetCf / getCurrentTarget / clearTarget tools once the server is running.
  • For production deployments, consider configuring secret management for credentials and enabling proper TLS termination on the gateway/exported URL.

Related MCP Servers

Sponsor this space

Reach thousands of developers