Skip to main content

MCP server for EdgescaleAI Cube operations

Project description

cube-mcp

MCP server for EdgescaleAI Cube management and Apollo deployments.

Install

npx cube-mcp

Requires uv or pipx.

Add to Claude Code

claude mcp add cube -- npx cube-mcp

Tools

Setup

Tool Description
setup_check Check all prerequisites and show status
setup_install_teleport Install Teleport CLI (tsh)
setup_install_kubectl Install kubectl
setup_install_helm Install Helm
setup_install_docker Install Docker
setup_install_apollo Guide for Apollo CLI installation
setup_configure_credentials Configure Apollo credentials

Cube Management

Tool Description
cube_login Login to a Cube (guided - shows SSO and password options)
cube_login_password Login via username/password (guided)
cube_list List available Cubes
cube_status Get Cube node status

App Proxy

Tool Description
app_list List available Teleport apps (HTTP/TCP)
app_login Login to a Teleport app
app_proxy Start local proxy for an app (MQTT, databases, etc.)
app_proxy_stop Stop running app proxies
app_proxy_status Show status of running proxies

Apollo Deployments

Tool Description
build_and_publish_to_apollo Build and publish app to Apollo
acr_login Login to Apollo Container Registry
acr_get_token Get ACR OAuth token
create_environment Create a new Apollo environment with control plane module
replicate_environment Clone an environment (modules, entities, configs, secrets)
install_entity Install a Helm chart entity on an environment
list_environments List Apollo environments (optionally filter by name)
entity_health Get health/activity status of entities in an environment
plan_details Deep-dive into entity plan details (tasks, events, error logs)

Quick Start

You: "Check my setup"
You: "Install kubectl"
You: "Connect me to staging-int"
You: "What's the Cube status?"
You: "List available apps"
You: "Proxy mosquitto-dev so I can connect locally"
You: "Build and publish my app"
You: "Create a new environment called my-test"
You: "Replicate pep-onlogicmc610-6 into a new env called pep-clone"
You: "List my Apollo environments"
You: "Show entity health for staging-int"
You: "Why did the dozzle install fail on pep-4?"
You: "Show plan details for redpanda-connect-20 on pep-4"

Credentials

Teleport

Login is guided - the tool will show you the commands to run:

  • Okta SSO: tsh login --proxy=edgescaleai.teleport.sh --auth=okta
  • Username/Password: tsh login --proxy=edgescaleai.teleport.sh --user=<username>

Apollo

Either method works — the MCP server checks both:

Option 1: apollo-cli configure (recommended)

apollo-cli configure

This stores credentials in your apollo-cli profile. The MCP server reads them automatically.

Option 2: Environment variables

export APOLLO_CLIENT="<your-client-id>"
export APOLLO_SECRET="<your-client-secret>"

Add to ~/.bashrc or ~/.zshrc and restart your terminal.

Local Development

First-time setup

# Clone the repo
git clone https://github.com/EdgescaleAI/cube-mcp.git
cd cube-mcp

# Create virtual environment and install
python3 -m venv .venv
source .venv/bin/activate
pip install -e .

# Add to Claude Code (use full path to venv python)
claude mcp add cube-local -- $(pwd)/.venv/bin/python -m cube_mcp.server

Reload after code changes

In Claude Code, run:

/mcp

This reloads all MCP servers and picks up your latest code changes.

Run server directly (for debugging)

source .venv/bin/activate
python -m cube_mcp.server

Documentation

See docs.md for detailed tool parameters and usage examples.

Contributing

  1. Create a branch
  2. Make changes
  3. Test locally with /mcp reload
  4. Push and create PR
  5. Merge to main → auto-publishes to PyPI and npm

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

cube_mcp-0.3.6.tar.gz (56.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

cube_mcp-0.3.6-py3-none-any.whl (27.0 kB view details)

Uploaded Python 3

File details

Details for the file cube_mcp-0.3.6.tar.gz.

File metadata

  • Download URL: cube_mcp-0.3.6.tar.gz
  • Upload date:
  • Size: 56.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for cube_mcp-0.3.6.tar.gz
Algorithm Hash digest
SHA256 29cfcfaf474fc54fc99eb514f055def07ac03101a98b4e9505a6aa7a8d3f3c5f
MD5 68cee0fbd7202d76af263da41418af82
BLAKE2b-256 6708c1de5a403d6164e68e90127eac54a866fbb235373789f9b420f566b5380b

See more details on using hashes here.

File details

Details for the file cube_mcp-0.3.6-py3-none-any.whl.

File metadata

  • Download URL: cube_mcp-0.3.6-py3-none-any.whl
  • Upload date:
  • Size: 27.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for cube_mcp-0.3.6-py3-none-any.whl
Algorithm Hash digest
SHA256 afdc8cc0b4840c4f22f60a912ee5a4a3bd6db11a7f80d4e9c35fdd04aeb6b73e
MD5 3d3e4ae12392d39398bb4ede6ba35215
BLAKE2b-256 630e38d48e965a4b354f6a6a106a298eee0203fb9ada185904bf1e6e574ab048

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page