spec-bug

Gather evidence from multiple sources and generate an investigation SPEC.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/jacobrocks1212/claude-config --skill spec-bug
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec-bug
Source: https://github.com/jacobrocks1212/claude-config/tree/main/user/skills/spec-bug
Command: npx skills add https://github.com/jacobrocks1212/claude-config --skill spec-bug

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It provides a structured workflow to investigate complex bugs by gathering evidence from multiple sources, confirming symptoms interactively with the user, and producing an INVESTIGATION SPEC to guide remediation.

Core Features & Use Cases

  • Parallel evidence gathering from conversation history, recent activity, and related documentation to ground the investigation.
  • Interactive symptom verification with the user to distinguish observed vs. expected behavior.
  • Generation of a formal SPEC that documents findings, hypotheses, open questions, and a plan for transitioning to a fix.

Quick Start

Describe the bug and request an evidence-led investigation SPEC.

Frequently Asked Questions about spec-bug

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

FAQPage Schema
How do I investigate a complex software bug and document the findings?

To investigate a complex software bug, gather evidence from conversation history, version changes, and documentation, then verify symptoms interactively to produce a formal investigation SPEC documenting hypotheses and a remediation plan.

What is an investigation SPEC and when do I need one for bug remediation?

An investigation SPEC is a structured document that captures evidence, confirmed symptoms, hypotheses, and open questions for a complex bug. You need one to guide remediation and transition logically into a fix plan.

How do I verify bug symptoms against expected behavior before planning a fix?

You verify bug symptoms through interactive confirmation, distinguishing observed behavior from expected behavior based on gathered evidence from code and documentation before finalizing a fix plan.

Can I use parallel evidence gathering to document complex code bugs?

Yes, you can use parallel subagents to collect evidence simultaneously from conversations, recent activity, documentation, and code, synthesizing the data into confirmed hypotheses for your bug documentation.

What is the best way to transition from bug investigation to a fix plan?

The best way to transition to a fix plan is by generating a structured SPEC that synthesizes gathered evidence and verified symptoms into actionable hypotheses and open questions for user confirmation.