ftm-debug

Orchestrates parallel agents to diagnose and verify codebase issues.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/kkudumu/feed-the-machine --skill ftm-debug
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ftm-debug
Source: https://github.com/kkudumu/feed-the-machine/tree/main/ftm-debug
Command: npx skills add https://github.com/kkudumu/feed-the-machine --skill ftm-debug

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep debugging of stubborn bugs by orchestrating a war room of parallel agents that instrument, research, reproduce, hypothesize, solve, and verify across the codebase.

Core Features & Use Cases

  • Parallel agent teams: Instrumenter, Researcher, Reproducer, Hypothesizer, Solver, and Reviewer work in concert in isolated worktrees.
  • Cross-cutting verification: automated reproduction, hypotheses testing, and evidence gathering before presenting results.
  • End-to-end debugging for complex, multi-turn issues that resist standard debugging.

Quick Start

Start a war-room debugging session to orchestrate parallel agents that instrument, research, reproduce, hypothesize, solve, and verify across your codebase.

Frequently Asked Questions about ftm-debug

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

FAQPage Schema
How do I debug complex bugs that resist standard debugging approaches?

Multi-agent debugging orchestrates parallel teams to instrument, research, reproduce, hypothesize, solve, and verify stubborn bugs across isolated worktrees. This coordinated war-room approach tackles complex, multi-turn issues that standard debugging cannot easily resolve.

What is a multi-agent debugging war room and how does it work?

A multi-agent debugging war room deploys specialized parallel agents—Instrumenter, Researcher, Reproducer, Hypothesizer, Solver, and Reviewer—working in concert across isolated worktrees. They cross-reference findings and enforce independent review before presenting verified results.

How do I start a parallel agent debugging session for my codebase?

Start a war-room debugging session to deploy parallel agents that instrument, research, reproduce, hypothesize, solve, and verify across your codebase. The session automates evidence gathering and verification gates before presenting results.

Does automated verification and hypothesis testing work for stubborn codebase bugs?

Automated verification and hypothesis testing enforce cross-cutting checks by reproducing bugs, testing hypotheses, and gathering evidence in isolated worktrees. This ensures results are validated before presentation, specifically targeting complex multi-turn issues.

When should I use coordinated multi-vector debugging instead of standard debugging?

Coordinated multi-vector debugging is designed for stubborn bugs that resist standard debugging approaches. It deploys parallel agents across your codebase to handle complex, multi-turn issues requiring end-to-end instrumented verification.

Does ftm-debug require any specific dependencies or external components to run?

ftm-debug operates with no dependencies or external components. It orchestrates a multi-agent debugging session entirely within isolated worktrees, using internal agents for instrumentation, reproduction, and verification.