debugging-workflows

Diagnose GitHub Agentic Workflow issues by analyzing logs and auditing runs.

Updated Feb 18, 2026
One-click install
npx skills add https://github.com/az9713/gh-aw --skill debugging-workflows-az9713
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debugging-workflows
Source: https://github.com/az9713/gh-aw/tree/main/.github/skills/debugging-workflows
Command: npx skills add https://github.com/az9713/gh-aw --skill debugging-workflows-az9713

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Clear, practical guidance for diagnosing and fixing issues in GitHub Agentic Workflows by analyzing logs, auditing runs, and understanding agent behavior.

Core Features & Use Cases

  • Log inspection and artifact analysis for failed runs.
  • Run auditing to identify failing steps and tool usage.
  • Debug agent interactions and guardrail outcomes across workflows.

Quick Start

Begin debugging by auditing a run and inspecting the most recent workflow logs with gh aw audit <run-id> and gh aw logs.

Frequently Asked Questions about debugging-workflows

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

FAQPage Schema
How do I debug a failing GitHub Actions workflow run?

Diagnose failing GitHub Agentic Workflows by running `gh aw audit <run-id>` to audit run behavior and `gh aw logs` to download logs, parse artifacts, and identify failing steps with structured guidance and safe defaults.

How do I inspect logs and artifacts for a failed GitHub Actions run?

Use the `gh aw logs` command to download and inspect workflow logs and parse artifacts, which helps you understand agent tool usage and identify the exact failing steps in your agentic workflows.

What is the best way to troubleshoot unexpected agent behavior in GitHub Actions?

The best way to troubleshoot unexpected agent behavior is to audit the GitHub Actions run to investigate agent tool usage and guardrail outcomes, providing structured guidance to debug interactions across workflows.

Can I analyze agent tool usage and guardrail outcomes in GitHub Agentic Workflows?

Yes, you can audit GitHub Agentic Workflow runs to analyze agent tool usage and debug guardrail outcomes, allowing you to investigate agent behavior and interactions across your workflow runs.

What do I need to start troubleshooting agentic workflows in GitHub Actions?

To start troubleshooting agentic workflows, you need the run ID of the target workflow to execute audit commands and access to GitHub Actions logs, allowing you to diagnose and fix issues using safe defaults.