architect-diagnose-performance

What problem does it solve?

It helps teams quickly diagnose why latency, errors, or throughput are regressing so they can prevent incidents and restore capacity.

Core Features & Use Cases

• Regression diagnosis across signals: Correlates latency/error distributions, user impact trends, machine health/usage, and recent commits or PRs to find likely causes.
• Bottleneck localization: Compares baseline vs incident conditions to identify the most affected route/service and the bottleneck stage driving the change.
• Safe remediation planning: Produces a remediation plan focused on the smallest safe intervention while capturing metrics and confidence.

Quick Start

Diagnose the current performance regression by identifying the worst affected route and incident window, comparing baseline vs incident metrics to infer the bottleneck stage, and proposing the smallest safe remediation.