debug

Diagnose errors and regressions using local command outputs, code, and external evidence.

18|8|Updated Jan 20, 2026
One-click install
npx skills add https://github.com/sydasif/web-search-mcp --skill debug-sydasif
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debug
Source: https://github.com/sydasif/web-search-mcp/tree/main/skills/debug
Command: npx skills add https://github.com/sydasif/web-search-mcp --skill debug-sydasif

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Debugging is a complex process where developers need to move from failing commands to root causes using a local-first workflow that combines Context7, GitHub CLI, and Web Search MCP to gather evidence and external references.

Core Features & Use Cases

  • End-to-end debugging workflow that starts from exact command outputs and local code to identify root causes.
  • Evidence-guided triage using Context7 for docs, gh for repo metadata, and Web Search MCP for external sources.
  • Use Case: Reproduce flaky tests, trace regressions, and verify fixes across environments.

Quick Start

Reproduce the error locally, capture the full command output and environment, and consult Context7, GitHub CLI, and Web Search MCP for evidence.

Frequently Asked Questions about debug

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

FAQPage Schema
How do I debug a failing command using local evidence?

Debugging a failing command involves reproducing the error locally and capturing the exact command output. You then gather evidence from local code, tests, and external signals using Context7, GitHub CLI, and Web Search MCP to isolate the root cause.

What is the best way to trace a regression in my local repository?

Tracing a regression locally requires capturing the exact command output and consulting git history. You use Context7 for documentation, GitHub CLI for issue history, and Web Search MCP for external references to validate the root cause and fix.

Can I use GitHub CLI and Web Search MCP to reproduce flaky tests?

Yes, you can reproduce flaky tests by running them locally to capture exact outputs. GitHub CLI provides repo metadata and issue history, while Web Search MCP fetches external references to help isolate environmental factors causing the flakiness.

Do I need a local git repository to diagnose errors with this workflow?

Yes, a local git repository is required to diagnose errors. The workflow is local-first, relying on local code, tests, and exact command outputs to gather evidence and validate fixes before consulting external documentation surfaces.

How does evidence-based debugging work with Context7?

Evidence-based debugging with Context7 works by gathering factual signals from documentation to support root cause analysis. You start with exact local command outputs, then query Context7 for docs to validate hypotheses and guide your fix.

Why is my debugging workflow not isolating the root cause effectively?

Your debugging workflow may fail to isolate the root cause if you skip capturing exact command outputs or neglect external signals. Combining local code inspection with GitHub CLI issue history and Web Search MCP references ensures comprehensive evidence gathering.