ferrox-debug

Diagnose broken software workflows with checkpoint tracking and isolated subagents.

21|9|Updated Jul 21, 2026

Search Tags:#debugging #root cause analysis #subagents #session management #workflow continuation #incident recovery

Ready to run

npx skills add https://github.com/FerroxLabs/ferrox-factory --skill ferrox-debug

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ferrox-debug
Source: https://github.com/FerroxLabs/ferrox-factory/tree/main/skills/ferrox-debug
Command: npx skills add https://github.com/FerroxLabs/ferrox-factory --skill ferrox-debug

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps when something is broken, failing, crashing, or behaving unexpectedly and you need a structured way to investigate it without losing progress after a context reset.

Core Features & Use Cases

Scientific debugging workflow: Investigates symptoms methodically instead of guessing at fixes.
Isolated subagent execution: Uses dedicated debug subagents to reduce confusion and keep work organized.
Checkpoint and continuation handling: Supports listing, resuming, and tracking active debug sessions across interruptions.
Use case: If an agent keeps producing incorrect output or a workflow fails midway, this Skill can diagnose the root cause, preserve state, and continue the investigation safely.

Quick Start

Use the ferrox-debug skill to investigate this failing workflow and return the root cause or apply a fix as appropriate.

Frequently Asked Questions about ferrox-debug

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

FAQPage Schema

How do I debug an agent workflow that keeps failing midway?▼

Debugging an agent workflow that fails midway requires a structured, scientific approach to investigate symptoms methodically rather than guessing. This skill diagnoses the root cause of failures and preserves investigation state safely across interruptions.

How do I resume a debugging session after a context loss?▼

To resume a debugging session after context loss, you need checkpoint tracking and continuation handling to list, track, and restore active sessions. This preserves your investigation state and allows you to continue diagnosing safely.

What is the best way to find the root cause of a crash without guessing?▼

The best way to find the root cause of a crash without guessing is applying a scientific debugging workflow that investigates symptoms methodically. It uses isolated subagents to reduce confusion and returns a structured diagnosis of the failure.

Can I run a diagnosis-only report for a misbehaving software workflow?▼

Yes, you can generate a diagnosis-only report for a misbehaving software workflow. The debugging process supports optional diagnosis-only reporting, returning the root cause analysis without automatically applying a fix.

Does isolated subagent orchestration help with incident recovery?▼

Isolated subagent orchestration helps with incident recovery by using dedicated debug subagents to keep work organized and reduce confusion. This ensures the investigation remains focused when diagnosing failures and preserving state across interruptions.