Skip to content
Google Cloud Observability MCP Server

Google Cloud Observability MCP Server

by Google
Released
8 views
v0.2.3
Apache-2.0
Repository Docs npm
Developer ToolsCloud & Infrastructure

Summary

Official Google server exposing Google Cloud Observability to agents — querying logs, metrics and traces from a Google Cloud project without leaving the assistant.

Features

  • Queries Google Cloud Logging, Monitoring and Trace through one MCP server
  • Pairs with the gcloud MCP server: telemetry for diagnosis, gcloud for action
  • Local stdio server run with npx, using your own Google Cloud credentials
  • Installable as a Gemini CLI extension
  • Apache 2.0, maintained by Google in the googleapis organisation

Installation

Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.

Any MCP-compatible agent

Most agents (Claude, Cursor, Windsurf, VS Code, and more) read a standard mcpServers configuration.

  1. Open your agent's MCP configuration file.
  2. Merge the snippet below into it.
  3. Restart the agent — the "Google Cloud Observability MCP Server" tools will be available.
{
  "mcpServers": {
    "google-cloud-observability-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@google-cloud/observability-mcp"
      ]
    }
  }
}

Description

Debugging a Google Cloud workload usually means three consoles: Cloud Logging for what was written, Cloud Monitoring for what the metrics did, and Cloud Trace for where the latency went. This official Google MCP server puts all three behind an agent, so an investigation can move between them in one conversation rather than three browser tabs.

The server talks to the Google Cloud Observability APIs directly, which makes it complementary to the sibling gcloud MCP server: gcloud is for changing and inspecting your environment, observability-mcp is for asking what that environment actually did. In practice the pairing is how most incident work goes — read the telemetry, form a hypothesis, then run the command.

Setup

It runs as a local stdio server, so any MCP client takes a standard entry running npx -y @google-cloud/observability-mcp. Node.js 20 or higher is required, along with Google Cloud credentials that carry the relevant observability read permissions on the project you intend to query. As with the other servers in the repository, it can be installed as a Gemini CLI extension using the package's init command.

Scope and licensing

Published in the same googleapis/gcloud-mcp repository as the gcloud, Cloud Storage and Backup and DR servers, under Apache 2.0. Google also runs a set of remote, fully managed MCP servers for Google Cloud products that authenticate over HTTP with your Google credentials; the local npm servers are the option to reach for when you want the agent talking to your own credentials from your own machine.

Related MCP Servers

MCP: ripwire

by Red Hat

New

Red Hat's zero-dependency C++23 code-context engine — ranked call graphs and blast-radius analysis, indexing a repo in under half a second with no server and no database.

Developer Tools
1 views

MCP: Graft

by Trail

New

Builds a searchable markdown graph of your repo so coding agents stop re-exploring it on every task — reported 42% fewer tokens and 46% fewer tool calls.

Developer Tools
1 views

MCP: Expo MCP Server

by Expo

New

Expo's official remote MCP server — searches Expo docs, installs compatible SDK packages, triggers and monitors EAS builds, and drives iOS/Android simulators.

Developer Tools
1 views

MCP: Hugging Face

by Hugging Face

New

Hugging Face's official MCP server — search the Hub's models, datasets and Spaces, read repository files, and call thousands of Gradio applications as tools from one remote endpoint.

Developer Tools
5 views
Browse all MCP servers →