Skill UI

Statistical gate for agent, prompt, model and tool changes: repeated trials, per-case paired comparison against a baseline, trajectory grading, flake detection and blocking exit codes before shipping.

This skill covers building, extending, and reviewing Model Context Protocol (MCP) servers using the Python FastMCP library. It includes guidance on managing API versions, defining tools, resources, and prompts, handling errors, and configuring transports. It also details how to evaluate agent workflows and test server capabilities against client requirements.