diagnose

Diagnose empirical results from CSV logs and CI outputs to identify patterns and root causes.

Updated Apr 29, 2026
One-click install
npx skills add https://github.com/smanist/a-exp --skill diagnose-smanist
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: diagnose
Source: https://github.com/smanist/a-exp/tree/main/.agents/skills/diagnose
Command: npx skills add https://github.com/smanist/a-exp --skill diagnose-smanist

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Diagnoses empirical results to identify patterns, root causes, and interpretation challenges, enabling teams to understand what the data really means.

Core Features & Use Cases

  • Identify systematic error patterns across experiments and logs.
  • Formulate testable root-cause hypotheses and plan subsequent actions.
  • Assess validity (construct, statistical, external, and ground-truth) of results and recommend next steps.

Quick Start

Run a diagnostic pass on your results set to produce an actionable diagnosis.

Frequently Asked Questions about diagnose

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I diagnose root causes in experiment results and CSV logs?

You can diagnose root causes in experiment results and CSV logs by running a structured workflow that formulates testable hypotheses, evaluates statistical validity, and identifies systematic error patterns to clarify what the empirical data means.

What is the best way to interpret empirical results from product experiments?

The best way to interpret empirical results from product experiments is by assessing construct, statistical, external, and ground-truth validity to identify patterns and generate recommended actions for subsequent testing steps.

Can I use this workflow to analyze CI outputs and performance metrics?

Yes, you can analyze CI outputs and performance metrics by diagnosing empirical results to identify systematic error patterns, formulate root-cause hypotheses, and generate recommended next steps across validation pipelines.

How do I assess statistical validity when diagnosing data interpretation issues?

You assess statistical validity when diagnosing data interpretation issues by evaluating construct, external, and ground-truth validity to identify systematic errors and formulate testable root-cause hypotheses.

Does diagnosing empirical results require any specific data formats or dependencies?

Diagnosing empirical results requires no specific dependencies and accepts CSV logs, experiment results, performance metrics, and CI outputs across studies, product experiments, and validation pipelines.