A Python package and MCP server for anytime-valid statistical checks on metrics, including A/B tests, change detection, denominator-aware rate checks, forecasting, and multiple-metric selection. It is designed to detect real signal versus sampling noise when data are peeked at repeatedly or stopped early.
From the registry: Is this number real, or is it noise? Peek-safe A/B tests, change detection, honest forecasts.
$ pip install noisefloorPlease install the `noisefloor` MCP server into my current AI client (that's you).
Optional prerequisites:
- Install Python package manager/runtime tooling — Install Python and a tool capable of running the MCP server entrypoint. The README shows installation via pip and MCP usage via uvx. Run: `pip install noisefloor`
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["--from","noisefloor","noisefloor-mcp"]
Note: README provides MCP configuration using the stdio transport via uvx. No environment variables are documented.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Trending hip-hop artist momentum scores across four cultural dimensions.
Official PostHog MCP Server for product analytics, feature flags, experiments, and more.
Official OpenMetadata MCP: governed context and business semantics for AI assistants and agents.