debug

Apply a repeatable debugging playbook to AI workflow issues.

4|2|Updated Aug 27, 2025
One-click install
npx skills add https://github.com/finger-gun/sisu --skill debug-finger-gun
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debug
Source: https://github.com/finger-gun/sisu/tree/main/packages/skills/skill-debug
Command: npx skills add https://github.com/finger-gun/sisu --skill debug-finger-gun

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Debugging AI workflows can be time-consuming and error-prone; this skill provides a structured, repeatable approach to reproduce issues, collect signals, isolate root causes, and verify fixes.

Core Features & Use Cases

  • Structured debugging playbook: step-by-step workflow from reproduction to root cause analysis.
  • Signal collection guidance: identify logs, traces, and reproduction steps to speed up triage.
  • Verification & documentation: capture fixes and regression checks to prevent reoccurrence.

Quick Start

Use the debug skill to apply a reproducible debugging workflow to your failing AI-powered process.

Frequently Asked Questions about debug

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

FAQPage Schema
What is a structured debugging playbook for AI workflows?

A debugging playbook for AI workflows is a repeatable process to identify and fix reproducible issues by collecting signals, isolating failure points, and verifying fixes across projects. It provides guided analysis for root-cause triage.

How do I isolate the root cause of a failing AI workflow?

To isolate the root cause of a failing AI workflow, follow a structured debugging playbook that guides you through signal collection, such as logs and traces, and systematic isolation of failure points before verifying fixes.

Can I use a debugging playbook for non-reproducible AI workflow issues?

A debugging playbook for AI workflows specifically applies to reproducible issues. Non-reproducible issues fall outside its primary scope, which focuses on capturing consistent reproduction steps and signals for guided analysis.

What is the best way to verify fixes and prevent regressions in AI workflows?

The best way to verify fixes in AI workflows is to apply a structured debugging playbook that includes regression checks and documentation, ensuring you capture the fix and prevent the issue from reoccurring across projects.

How do I collect signals to speed up triage for AI workflow troubleshooting?

To speed up troubleshooting in AI workflows, use a debugging playbook to identify and collect signals like logs, traces, and reproduction steps, allowing you to systematically isolate failure points and verify the fix.