What it does

Exposes Google Cloud Observability APIs through the Model Context Protocol, allowing AI assistants to query logs, metrics, and traces from your GCP environment using natural language. The server wraps Cloud Logging, Cloud Monitoring, and Cloud Trace behind 13 tools, eliminating manual API construction or complex CLI invocations for observability operations.

Who it's for

DevOps engineers and SREs investigating production incidents in GCP, or platform teams automating observability queries. Works with Claude, Cline, Cursor, and Gemini CLI—useful for teams that want to correlate logs, metrics, and traces without switching to the GCP console.

Common use cases

Query Cloud Logging for error patterns and exceptions across services
Pull metrics from Cloud Monitoring to diagnose performance issues
Retrieve distributed traces from Cloud Trace to understand request paths
Aggregate observability data for incident postmortems

Setup pitfalls

Requires gcloud CLI installed and authenticated to your GCP project
Node.js 20 or higher is required
Target GCP project must have Observability APIs enabled and appropriate IAM roles for logging, monitoring, and trace read access
The server reads and may write files in the gcloud configuration directory; ensure filesystem permissions allow this

Tool name

Description

Destructive?

list_log_entries

✓ no

list_log_names

✓ no

list_buckets

✓ no

list_views

✓ no

list_sinks

✓ no

list_log_scopes		✓ no
list_metric_descriptors		✓ no
list_time_series		✓ no
list_alert_policies		✓ no
list_alerts		✓ no
list_traces		✓ no
get_trace		✓ no
list_group_stats		✓ no

@google-cloud/observability-mcp

What it does

Who it's for

Common use cases

Setup pitfalls