gsd-debug

Maintain persistent debugging sessions across context resets with checkpointed analysis.

4|1|Updated Mar 22, 2026
One-click install
npx skills add https://github.com/lgwanai/openforce --skill gsd-debug-lgwanai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gsd-debug
Source: https://github.com/lgwanai/openforce/tree/main/skills/gsd-debug
Command: npx skills add https://github.com/lgwanai/openforce --skill gsd-debug-lgwanai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Debugging non-deterministic or hard-to-reproduce issues becomes slow when context resets or you lose intermediate findings, causing you to re-investigate the same symptoms repeatedly.

Core Features & Use Cases

  • Persistent, checkpointed debugging sessions: preserves investigation state across context resets so you can resume instead of starting over.
  • Scientific-method style isolation: gathers symptoms, spawns a dedicated debugger subagent, and narrows toward root cause through structured investigation.
  • Diagnose-only mode: uses --diagnose to return a Root Cause Report without applying changes, suitable for risk-controlled verification.
  • Session control by slug: supports list, status <slug>, and continue <slug> to manage multiple debug threads.

Quick Start

Run openforce /gsd-debug --diagnose "Describe the issue symptoms you are seeing and what changed right before it started".

Frequently Asked Questions about gsd-debug

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

FAQPage Schema
How do I resume debugging a complex issue after a context reset without losing my findings?

Persistent, checkpointed debugging sessions preserve your investigation state across context resets. Use subcommands like `list`, `status <slug>`, and `continue <slug>` to manage and resume multiple debug threads without losing intermediate findings.

What is the best way to perform root cause analysis on hard-to-reproduce bugs?

Root cause analysis for non-deterministic issues uses scientific-method style isolation. The Skill gathers symptoms, spawns a dedicated debugger subagent, and narrows toward the root cause through structured investigation across multiple checkpoints.

How do I diagnose a bug and verify its root cause without applying fixes to my code?

Use the `--diagnose` flag to run diagnose-only mode. This performs the full investigation and returns a Root Cause Report without applying any changes, ensuring risk-controlled verification before you commit to fixes.

Can I manage multiple debugging sessions simultaneously for incident response?

Yes, you can manage multiple debugging threads simultaneously using session control by slug. The Skill supports `list`, `status <slug>`, and `continue <slug>` subcommands to track and resume distinct incident response investigations.

Does this debugging approach work for non-deterministic issues that take a long time to reproduce?

Yes, this approach specifically targets non-deterministic issues requiring persistent investigation. By checkpointing your analysis state, it prevents you from losing intermediate findings and avoids restarting the investigation when context resets occur.