reflect

Use when the user wants to analyze AI coding agent telemetry, generate a Reflect dashboard, investigate tool/model/MCP/subagent patterns, understand Claude, Copilot, or Gemini usage/limit behavior, or decide between local-first and central gateway-based telemetry flows. Start with local OTLP/span analysis, then optionally enrich the explanation with provider-side admin APIs, exports, or quota views when available.

Use when the user wants to extract reusable skills from their AI session history and distribute them to their coding agents.