# AI Tools and Skills

> Source: <https://signoz.io/docs/ai/overview>
> Published: 2026-05-27 00:00:00+00:00

SigNoz provides two ways to connect AI coding assistants to your observability data, allowing you to seamlessly set up SigNoz, write optimized queries, or analyze your telemetry:

Integration Options

MCP Server

The [SigNoz MCP Server](https://github.com/SigNoz/signoz-mcp-server) implements the Model Context Protocol (MCP), enabling direct communication between AI assistants and your SigNoz instance.

- Query metrics, logs, and traces using natural language
- Inspect alerts, dashboards, and services
- Works with Claude, Cursor, Gemini, Codex, etc.

[Get started with the MCP Server →](https://signoz.io/docs/ai/signoz-mcp-server)

Agent Skills

[SigNoz Agent Skills](https://github.com/SigNoz/agent-skills) are an open format for extending AI coding assistants with specialized knowledge about SigNoz. Skills are markdown files (`SKILL.md`

) that guide agents to perform observability tasks.

- Explore SigNoz documentation and ClickHouse query patterns
- Works with Claude Code and skills.sh compatible agents like Codex, Cursor, Gemini, OpenCode, etc.
- Install with a single command

[Get started with Agent Skills →](https://signoz.io/docs/ai/agent-skills)

MCP Use Cases

Once you have the MCP server connected, explore practical workflows:

- Search and analyze logs by asking questions in plain English.
- Ask "why is this slow?" and get a span breakdown with the bottleneck identified.
- Paste a trace ID and reconstruct the full request path with root cause analysis.