Helm Chart Cli

by jeff-nasseri

223 downloads Not rated yet
GitHub

About

Helm MCP provides a bridge between AI assistants and the Helm package manager for Kubernetes. It allows AI assistants to interact with Helm through natural language requests, executing commands like installing charts, managing repositories, and more.

Explore

- Chart creation, linting, packaging, and template rendering.
- Comprehensive release management: install, upgrade, rollback, uninstall, list, status, history, test.
- Dependency management: build, list, update.
- Repository management: add, remove, list, update, search.
- Registry management: login, logout, push, pull.
- Release information: get all, hooks, manifest, metadata, notes, values.

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

  1. Download and install Highlight from highlightai.com/download
  2. Navigate to the plugins tab and select "Add Custom Plugin"
  3. Configure the plugin with the settings below
    Plugin Name Helm Chart Cli
    Command (node, npx, python, etc.)

    Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.

  4. Enable "Start Automatically" if you want the plugin to start when Highlight launches

From the repository

- Python 3.8+
- Docker (for containerized deployment)
- Helm CLI installed


uv venv
source .venv/Scripts/Activate.ps1
uv pip install -e .

Installs a chart.
- Parameters:
- chart (required): Chart name.
- release_name (optional): Release name.
- namespace (optional): Namespace.
- values_file (optional): Values file.
- set_values (optional): Set values.
- description (optional): Add a custom description.
- wait (optional): Wait until all resources are ready.
- atomic (optional): If set, installation rollback on failure.
- timeout (optional): Time to wait for any operation to complete.
- Example:


helm_install(chart="bitnami/nginx", release_name="my-nginx")

Uninstalls a release.
- Parameters:
- release_name (required): Release name.
- namespace (optional): Namespace.
- keep_history (optional): Remove the release but keep the history.
- no_hooks (optional): Prevent hooks from running during uninstallation.
- Example:

helm_uninstall(release_name="my-nginx")

Installs a Helm plugin.
- Parameters:
- plugin_url (required): URL to the plugin.
- version (optional): Version of the plugin.
- Example:

helm_plugin_install(plugin_url="https://github.com/chartmuseum/helm-push")

Uninstalls a Helm plugin.
- Parameters:
- plugin_name (required): Name of the plugin.
- Example:

helm_plugin_uninstall(plugin_name="push")

Add this to your claude_desktop_config.json:

<details>
<summary>Using uvx</summary>

json
"mcpServers": {
"helm": {
"command": "uvx",
"args": ["mcp-server-helm"]
}
}
</details>

<details>
<summary>Using docker</summary>

json
"mcpServers": {
"helm": {
"command": "docker",
"args": ["run", "--rm", "-i", "mcp/helm"]
}
}
```
</details>

Here are the available tools in the Helm MCP server:

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "helm chart cli": {
            "helm": {
                "command": "docker",
                "args": [
                    "build",
                    "-t",
                    "mcp-helm",
                    "."
                ]
            }
        }
    }
}

McpServers

{
    "helm": {
        "command": "docker",
        "args": [
            "build",
            "-t",
            "mcp-helm",
            "."
        ]
    }
}

run the inspector against the mcp-server-helm

npx @modelcontextprotocol/inspector uvx mcp-server-helm

Build

Docker build:

bash
cd src/helm
docker build -t mcp/helm .
```

License

This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please see the LICENSE file in the project repository.

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.