debug-issue

Trace and debug code issues using knowledge graph call-chain navigation.

25|4|Updated Dec 21, 2025
One-click install
npx skills add https://github.com/thangchung/agent-engineering-experiment --skill debug-issue-thangchung
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: debug-issue
Source: https://github.com/thangchung/agent-engineering-experiment/tree/main/agentgateway-entraid-obo/.claude/skills/debug-issue
Command: npx skills add https://github.com/thangchung/agent-engineering-experiment --skill debug-issue-thangchung

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Debugging unfamiliar or large codebases often means hours of manual searching to understand how a bug propagates through call chains. This Skill uses a knowledge graph of the codebase to systematically trace issues from symptom to root cause. ## Core Features & Use Cases - Semantic Code Search: Find code related to an issue using semantic search over graph nodes instead of keyword grep. - Call Chain Tracing: Follow callers and callees to understand the full execution context around a suspected bug. - Change & Impact Analysis: Detect recent changes that may have introduced the issue and measure the impact radius of suspected files. - Use Case: A production endpoint started failing after a recent commit. Use this Skill to locate the relevant code, trace the execution flow to the entry point, and confirm which recent change caused the regression. ## Quick Start Debug the failing checkout flow by tracing its call chain and checking recent changes in the knowledge graph.

Frequently Asked Questions about debug-issue

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

FAQPage Schema
How do I debug a bug using a code knowledge graph?▼

Start with semantic search to find code related to the issue, then trace callers and callees to follow the call chain. Use flow retrieval to see full execution paths and change detection to check whether recent edits caused the problem.

How to trace function callers and callees in a codebase?▼

Use the graph query tool with callers_of and callees_of on the suspected function. Checking both directions reveals the full context, from the entry point that triggers the code down to the operations it invokes.

Can knowledge graph debugging find which commit caused a bug?▼

Yes, the change detection tool identifies recent modifications in the codebase. Since recent changes are the most common source of new issues, correlating them with the affected flow usually pinpoints the culprit.

How do I reduce token usage when using code graph tools?▼

Always begin with get_minimal_context for your task before calling other graph tools, and use detail_level minimal on all calls. Only escalate to standard detail when minimal output is insufficient, targeting five or fewer tool calls.

What is impact radius analysis in code debugging?▼

Impact radius analysis shows which other files and flows are affected by a suspected file. Running it on candidate files helps you understand the blast radius of a bug and verify whether a fix might break dependent code.