debugging

Debug software issues by analyzing monitoring data, logs, and version history.

Updated Mar 31, 2026
One-click install
npx skills add https://github.com/Saturate/agents --skill debugging-saturate
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: debugging
Source: https://github.com/Saturate/agents/tree/main/skills/debugging
Command: npx skills add https://github.com/Saturate/agents --skill debugging-saturate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Stops guessing by driving debugging with evidence gathered from monitoring, logs, and context, reducing wasted time on incorrect hypotheses.

Core Features & Use Cases

  • Evidence-driven debugging: checks monitoring/APM data, gathers context, localizes the issue to a layer, and records hypotheses and attempts to avoid loops.
  • Troubleshooting workflows: suitable for error investigation, finding root cause, and diagnosing issues in software systems.
  • Use Case: When facing a bug, you can follow the steps to read errors, verify monitoring data, gather context, localize to the layer, and verify fixes with regression checks.

Quick Start

Follow the evidence-based debugging workflow to read errors, verify monitoring data, gather context, localize the issue, and validate fixes.

Frequently Asked Questions about debugging

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

FAQPage Schema
What is evidence-driven debugging and how does it find root causes?▼

Evidence-driven debugging finds root causes by checking monitoring data and logs to validate stepwise hypotheses, stopping guesswork and reducing wasted time on incorrect theories during software troubleshooting.

How do I troubleshoot software errors using monitoring and log data?▼

Troubleshoot software errors by reading logs, verifying APM monitoring data, gathering context, localizing the issue to a specific layer, and validating fixes with regression checks.

How do I stop debugging loops and avoid repeating failed hypotheses?▼

Stop debugging loops by recording hypotheses and attempts throughout your investigation, ensuring you track stepwise progress and avoid repeatedly testing incorrect root-cause theories.

What data do I need to localize an issue to a specific system layer?▼

Localizing an issue to a specific system layer requires access to monitoring data, application logs, and version history to gather enough context for accurate root-cause identification.

Does this troubleshooting workflow work for diagnosing issues in any software system?▼

Yes, this troubleshooting workflow applies across error investigation, finding root causes, and diagnosing issues in software systems, provided you can access monitoring data and logs.

How do I validate a bug fix and prevent regressions during debugging?▼

Validate bug fixes and prevent regressions by enforcing regression checks after localizing the issue, ensuring your evidence-driven hypothesis resolution remains stable across the system.