The Polar Signals MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Polar Signals account. It provides structured and secure access to your continuous profiling projects and performance data, so your agent can list projects, explore profile types and labels, analyze flame graphs, inspect function relationships, and identify optimization opportunities on your behalf.
- Project and profile discovery: Have your agent list accessible projects and find the profile types available for performance analysis.
- Flame graph analysis: Let the agent query flame graphs as call trees to investigate where applications spend their time.
- Function performance ranking: Direct your agent to compare functions in a flat table sorted by cumulative time and spot likely optimization targets.
- Caller and callee inspection: Instruct your agent to examine which functions call a chosen function and which functions it calls.
- Label and source analysis: Have your agent explore profile labels and their values, then retrieve per-line profiling data for a specific source file.