Observability FAQs

Question 1

Which observability platforms does it integrate with?

Accepted Answer

It integrates with Grafana for Loki log queries (using LogQL) and with Prometheus or Thanos for metric queries (using PromQL). A combined server option offers unified access to both log and metric data.

Question 2

What types of queries can AI assistants perform using Observability?

Accepted Answer

AI assistants can perform a range of queries, including LogQL range and instant queries for Loki logs, and PromQL instant and range queries for Prometheus/Thanos metrics. They can also list available labels and time-series.

Question 3

What authentication methods does Observability support?

Accepted Answer

It offers robust authentication options including browser-cookie SSO (for Grafana), service-account tokens, static session cookies, and OIDC via kubeconfig, ensuring secure and flexible access to your data sources.

Question 4

What is Observability (observability-mcp)?

Accepted Answer

Observability is an MCP (Model Context Protocol) server designed to connect AI assistants directly to your observability stack. It enables AI clients to query logs from Grafana/Loki and metrics from Thanos/Prometheus programmatically.

Question 5

How does Observability connect with AI assistants?

Accepted Answer

Observability utilizes the Model Context Protocol (MCP) to expose your observability data to MCP-compatible AI clients, such as Cursor and Claude Desktop. This allows AI to understand and execute queries against your logs and metrics.

Observability

Observability

主な機能

ユースケース

主な機能

ユースケース