Tideways

You can query Tideways performance data with an MCP server that exposes Tideways metrics and insights to AI assistants. This server translates performance data into natural-language insights, supports real-time queries, and helps you analyze errors, traces, and optimization opportunities for your PHP applications.

How to use

Once you configure the MCP server in your AI assistant’s MCP settings, you can ask natural-language questions like “What’s the current performance of my application?” or “Show me the slowest transactions right now.” You can also request deeper analysis such as tracing a specific endpoint, identifying bottlenecks, or getting optimization recommendations. The server uses a stdio transport to communicate with compatible AI assistants, returning structured data that your assistant can render in context.

How to install

Prerequisites you need before you start:

• Node.js and npm installed on your machine.

Step 1: Install the MCP server globally (optional but convenient)

npm install -g tideways-mcp-server

Step 2: Or use npx to run without a global install (recommended for testing or quick starts)

npx tideways-mcp-server

Step 3: Prepare environment variables for your Tideways account

TIDEWAYS_TOKEN=YOUR_TOKEN
TIDEWAYS_ORG=YOUR_ORG
TIDEWAYS_PROJECT=YOUR_PROJECT

Step 4: Start the server with your MCP client configuration

TIDEWAYS_TOKEN=YOUR_TOKEN TIDEWAYS_ORG=YOUR_ORG TIDEWAYS_PROJECT=YOUR_PROJECT npx tideways-mcp-server

Configuration and usage notes

This MCP server is designed to work with AI assistants that support MCP via stdio. You provide the connection details in the assistant’s MCP configuration file and set the required environment variables when starting the server. The server exposes tools that return raw JSON data, which your AI assistant can present in conversational form.

Security and environment

Keep sensitive credentials secure by loading them from environment variables and never hard-coding tokens in source files. The server redacts authorization headers in logs and validates inputs for all MCP interactions.

Troubleshooting tips

If you encounter authentication errors, verify that TIDEWAYS_TOKEN, TIDEWAYS_ORG, and TIDEWAYS_PROJECT are correct and have the required scopes. If you hit rate limits, adjust your query frequency or rely on the server’s retry logic. For MCP connection issues, ensure your assistant’s MCP configuration is correct and that the server process is running with the proper environment variables.

Using with AI assistants

After configuration, ask your AI assistant questions about current performance, recent errors, or trace-level bottlenecks. The server provides structured JSON suitable for direct interpretation by the assistant, enabling natural-language reporting and actionable recommendations.

Available tools

get_performance_metrics

Retrieve aggregate performance metrics and system-wide statistics.

get_performance_summary

Retrieve time-series performance summary data in 15-minute intervals for trend analysis.

get_issues

Retrieve and analyze recent errors, exceptions, and performance issues.

get_traces

Analyze individual trace samples for bottlenecks and optimization recommendations.

get_historical_data

Retrieve historical performance data for specific dates with configurable granularity.