Mcp Server Gcp Manager
About
Mcp Server Gcp Manager is an open-source Model Context Protocol (MCP) server that enables AI agents (such as Claude, Llama 3.1, or custom LangChain bots) to interact with Google Cloud Platform safely and autonomously. It bridges an LLM with the gcloud CLI, allowing…
Details
- Author
- KishoreKu
- Downloads
- 184
- Categories
- Cloud Service
Jump to
- List Cloud Run services with detailed inspection.
- Create and update secrets in Google Secret Manager.
- Automatically enable the Secret Manager API if disabled.
- Deploy new Cloud Run revisions (update memory, image, etc.).
- Check existing configuration before applying changes to prevent accidental overwrites.
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:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Mcp Server Gcp ManagerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Installation requires Python 3.10+, a working gcloud installation authenticated with a project, and the mcp Python package. Optionally set the GCLOUD_PATH environment variable if gcloud is not auto‑detected. For Claude Desktop, add a gcp-manager entry to claude_desktop_config.json. For LangChain or custom agents, use langchain-mcp-adapters to connect to the server via stdio transport.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"mcp server gcp manager": {
"gcp-manager": {
"command": "python",
"args": [
"/path/to/mcp-server-gcp-manager/server.py"
],
"env": {
"GCLOUD_PATH": "/path/to/your/gcloud"
}
}
}
}
}McpServers
{
"gcp-manager": {
"command": "python",
"args": [
"/path/to/mcp-server-gcp-manager/server.py"
],
"env": {
"GCLOUD_PATH": "/path/to/your/gcloud"
}
}
}Google Cloud MCP Server
An open-source Model Context Protocol (MCP) server that enables AI agents (like Claude, Llama 3.1, or custom LangChain bots) to interact with Google Cloud Platform safely and autonomously.
This server acts as a bridge between your LLM and the gcloud CLI, allowing you to manage Cloud Run services and Secrets using natural language.
🚀 Features
- 🔍 List Services: detailed inspection of running Cloud Run services.
- 🔐 Manage Secrets: Safely create and update secrets in Google Secret Manager.
- Self-Healing: Automatically detects if the Secret Manager API is disabled and enables it for you.
- 🚀 Deploy & Update: Deploy new revisions or update memory/image configuration for Cloud Run services.
- Safe-Guards: Checks existing configuration before applying changes to prevent accidental overwrites.
🛠️ Prerequisites
1. Python 3.10+
2. Google Cloud SDK (gcloud): Must be installed and authenticated.
- Install: brew install --cask google-cloud-sdk (Mac) or see Official Docs.
- Auth: Run gcloud auth login and gcloud config set project [YOUR_PROJECT_ID].
📦 Installation
1. Clone the repository:
git clone https://github.com/YOUR_USERNAME/mcp-gcp-manager.git
cd mcp-gcp-manager
2. Create a Virtual Environment (Recommended):
python3 -m venv venv
source venv/bin/activate
3. Install Dependencies:
pip install mcp
⚙️ Configuration
The server attempts to automatically find your gcloud executable.
If you see a "gcloud not found" error: You can manually tell the server where gcloud is by setting an environment variable:
1. Example for Mac/Linux
export GCLOUD_PATH="/usr/local/bin/gcloud"
2. Example for Windows (PowerShell)
$env:GCLOUD_PATH="C:\Program Files\Google\Cloud SDK\google-cloud-sdk\bin\gcloud.exe"
💻 Usage
Option 1: Using with Claude Desktop
To use this with the official Claude Desktop app, add the following to your claude_desktop_config.json:
{
"mcpServers": {
"gcp-manager": {
"command": "/path/to/your/venv/bin/python",
"args": ["/path/to/mcp-gcp-manager/server.py"]
}
}
}
Option 2: Using with LangChain / Custom Python Agents
You can connect to this server using langchain-mcp-adapters:
from langchain_mcp_adapters.client import MultiServerMCPClient
3. Connect to the local server process
client = MultiServerMCPClient({ "gcp": { "command": "python", "args": ["server.py"], "transport": "stdio" } })4. Get tools and bind to your model (Llama 3, GPT-4, etc.)
tools = await client.get_tools() ``
🛡️ Tools Available
| Tool Name | Description |
| :--- | :--- |
|
list_running_services | Lists all Cloud Run services in a specific project. |
| create_secret | Creates/Updates secrets. Handles API enablement automatically. |
| deploy_service` | Deploys a new revision. Supports updating memory, image, etc. |
🤝 Contributing
Pull requests are welcome! Please ensure any new tools include proper error handling and do not hardcode local paths.
📄 License
MIT
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
