Signoz MCP Server acts as a crucial bridge, connecting AI assistants and Large Language Models to SigNoz's observability data. Developed by Signoz using Go 1.25 or higher, this free Model Context Protocol server enables natural language interaction with metrics, traces, and logs. It integrates with both SigNoz Cloud and self-hosted SigNoz instances, allowing developers, SREs, and platform teams to enhance their observability workflows directly within AI-driven environments. Its design supports various transport modes and authentication methods for flexible deployment.

What does Signoz MCP Server connect to?

The Signoz MCP Server connects directly to a SigNoz instance, functioning as an MCP server. This instance can be either a hosted SigNoz Cloud deployment or a self-hosted SigNoz environment. Its primary role is to access existing observability data within SigNoz. It serves as an intermediary component, effectively bridging the gap between AI assistants or LLMs and the rich observability data stored and managed by SigNoz, enabling external systems to query and interact with this information.

Core features and tools

The Signoz MCP Server exposes a defined set of callable tools to interact with observability data. These include Metrics Exploration for querying performance data, Alert Monitoring for managing notifications, and Dashboard Management for interacting with visualizations. Further capabilities cover Service Analysis for understanding application behavior, Log Analysis for examining events, and Trace Analysis for distributed transaction insights. Additionally, it offers a specialized tool, signoz_fetch_doc, for retrieving documentation directly through the protocol.

Who is it for?

The Signoz MCP Server is designed for developers, Site Reliability Engineers (SREs), and platform teams. Its utility is particularly high for those who employ AI assistants and Large Language Models to aid in observability tasks. This includes individuals needing to interact with metrics, traces, logs, and alerts using natural language queries. Skip this if you do not use AI assistants or LLMs as part of your observability practice.

How to get started

For SigNoz Cloud users, access is granted through a standard OAuth authorization flow, utilizing Authorization Code with PKCE, where clients dynamically register and obtain tokens. For self-hosted instances, the server authenticates using an API key passed as an environment variable, specifically "SIGNOZ_API_KEY", along with "SIGNOZ_URL", with communication via "stdio" (default) or "http" transport.

Use cases

• Querying metrics, traces, logs, alerts, and dashboards with natural language commands, simplifying data retrieval.
• Debugging production issues by providing rich observability context including traces, logs, metrics, service topology, and deployment history directly within coding agents.
• Investigating incidents, explaining complex dashboards, and exploring detailed data without requiring users to master intricate query languages.

The Signoz MCP Server serves as a vital link, empowering AI assistants and LLMs to interact with SigNoz observability data. It provides structured access to key areas like metrics, logs, and traces, supporting both cloud and self-hosted deployments with distinct authentication mechanisms. This server is a valuable asset for teams aiming to integrate AI into their operational workflows, facilitating natural language queries and incident investigation. Try querying metrics, traces, or logs using natural language through a compatible AI client.