debug-auto

Coordinate automated bug investigation, fix, and verification via dispatched sub-agents.

4|2|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/blake-simpson/belmont --skill debug-auto
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debug-auto
Source: https://github.com/blake-simpson/belmont/tree/main/plugin/skills/debug-auto
Command: npx skills add https://github.com/blake-simpson/belmont --skill debug-auto

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Auto-debug loops are tedious and error-prone when managed manually; this Skill coordinates an end-to-end investigation, fix, and verification cycle via dispatched agents, reducing time and human effort.

Core Features & Use Cases

  • Orchestrates an investigate-fix-verify loop using design, implementation, and verification agents
  • Manages a three-iteration cycle with iteration tracking and centralized coordination in DEBUG.md
  • Provides a safe, auto-mode workflow for small bug fixes without feature work

Quick Start

Start a debug-auto session and let Belmont orchestrate a three-iteration auto-debug loop using dispatched agents.

Frequently Asked Questions about debug-auto

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

FAQPage Schema
How do I automate an auto-debug loop for fixing small bugs?

You can automate an auto-debug loop by dispatching sub-agents through a three-phase investigate, fix, and verify cycle. The loop runs up to three iterations, tracking progress and logging results centrally in DEBUG.md.

What is the best way to orchestrate bug investigation and resolution with agents?

Agent orchestration for bug resolution coordinates sub-agents across design, implementation, and verification phases. It manages a three-iteration cycle with centralized coordination logged in DEBUG.md to reduce manual effort.

Can I use automated debugging agents for large feature work?

Automated debugging agents are designed for small, scoped issues suitable for auto-mode only. They specifically avoid large feature work, focusing instead on end-to-end investigation, fix, and verification cycles for minor bugs.

How does the three-iteration cap work in an automated debug cycle?

The three-iteration cap limits the automated debug cycle to three sequential passes of design, implementation, and verification. Each iteration is tracked centrally in DEBUG.md to coordinate sub-agent dispatches and report results.

Why does my manually managed auto-debug workflow fail to verify fixes?

Manually managed auto-debug loops are tedious and error-prone due to lack of centralized coordination. Automating the investigate-fix-verify cycle with dispatched agents ensures structured verification and tracks results in DEBUG.md.