signoz

This MCP server exposes Signoz-related capabilities as tools that you can call from an AI assistant or your own client. Available tools include: test_connection to verify connectivity to your Signoz instance, fetch_dashboards to list dashboards, fetch_dashboard_details to retrieve metadata for a specific dashboard by ID, fetch_dashboard_data to pull all panel data for a dashboard within a time range, fetch_apm_metrics for standard APM metrics by service and time range, fetch_services to enumerate instrumented services, execute_clickhouse_query to run customized ClickHouse SQL queries via the Signoz API with time support, execute_builder_query to run Signoz builder queries for custom metrics, and fetch_traces_or_logs to retrieve traces or logs (specifying data_type as 'traces' or 'logs') with time range, service name, and limit. To use these tools, point your MCP-enabled assistant at the Signoz MCP server (via HTTP/SSE) and invoke the desired tool names along with any required parameters (such as dashboard_id, service_name, time_range, or query). The server supports both local and remote deployments and can be run in development with uv or in production via Docker Compose. Ensure your Signoz host and optional API key are configured correctly (via environment variables or YAML config) before calling the tools.

Prerequisites:

• Docker or Python environment (depending on your deployment choice)
• Access to a Signoz instance (self-hosted or cloud) with host URL and optional API key
• Optional: Docker Compose for containerized deployment

Install & Run Options:

2A. Install & Run with uv (Recommended for Local Development)

1. Install dependencies and create virtual environment:

uv venv .venv
source .venv/bin/activate
uv sync

2. Run the MCP server:

uv run -m src.signoz_mcp_server.mcp_server

Note: Place config.yaml in the same directory as mcp_server.py or set corresponding environment variables as described in the Configuration section.

2B. Run with Docker Compose (Recommended for Production/Containerized Environments)

1. Edit src/signoz_mcp_server/config.yaml with your Signoz details (host, API key if needed).
2. Start the server:

docker-compose up -d

The server will run in HTTP (SSE) mode on port 8000 by default. You can override configuration with environment variables (see Configuration section).

Prerequisites for Docker-based deployment:

• Docker and Docker Compose installed on your host
• Access to the Signoz instance URL and optional API key
• Proper network access from the host to Signoz

Tips and common notes:

• The MCP server defaults to port 8000; override with MCP_SERVER_PORT if needed.
• Configuration can be provided via environment variables (SIGNOZ_HOST, SIGNOZ_API_KEY, SIGNOZ_SSL_VERIFY) or a YAML file (config.yaml).
• When using Docker, the -t stdio flag helps MCP clients communicate via standard input/output.
• If you encounter connectivity issues, verify that SIGNOZ_HOST is reachable from the MCP server host and that SSL verification settings match your Signoz deployment.
• For local testing, you can use http://localhost:8000/mcp as the endpoint in your MCP client configuration.
• Health check: curl http://localhost:8000/health