skills.rest
WeWake1
WeWake1COMMUNITY
Explore all skills by this author →
GitHub

gsd-debug

Debug issues systematically using subagent isolation with persistent session state.

Updated Mar 15, 2026
Search Tags:#debugging#context management#troubleshooting#root cause analysis#subagent#session persistence#scientific method
One-click install
npx skills add https://github.com/WeWake1/BSOMS --skill gsd-debug-wewake1
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: gsd-debug
Source: https://github.com/WeWake1/BSOMS/tree/main/.agent/skills/gsd-debug
Command: npx skills add https://github.com/WeWake1/BSOMS --skill gsd-debug-wewake1

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Debugging complex issues consumes large amounts of context when reading files, forming hypotheses, and testing fixes, leaving little room for user interaction. This Skill offloads investigation to isolated subagents while persisting session state across context resets. ## Core Features & Use Cases - Structured Symptom Gathering: Collects expected behavior, actual behavior, error messages, timeline, and reproduction steps before investigation begins. - Subagent-Based Investigation: Spawns a gsd-debugger agent with a fresh context window to investigate, keeping the main conversation lean. - Persistent Debug Sessions: Stores session state in .planning/debug/ markdown files so investigations survive context resets and can be resumed later. - Checkpoint Handling: Supports human-verify checkpoints where the user confirms fixes before the agent finalizes and archives the session. - Use Case: A developer hits a failing build after a refactor. The Skill gathers symptoms, spawns a debugger agent that isolates the root cause, presents evidence, and offers to fix it immediately or plan the fix. ## Quick Start Ask the assistant to debug your issue by describing the unexpected behavior, and it will gather symptoms and launch an isolated investigation.

Frequently Asked Questions about gsd-debug

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

FAQPage Schema
How do I debug an issue without losing conversation context?▼

Use subagent-based debugging where a fresh agent with its own context window handles file reading and hypothesis testing. The main conversation only coordinates symptom gathering, checkpoints, and user decisions.

How do I resume a previous debugging session?▼

Run the debug command without arguments to list active sessions stored in .planning/debug/. Pick a session number to resume, and a continuation agent reloads the saved state from the session's markdown file.

What information should I gather before debugging a bug?▼

Collect five items: expected behavior, actual behavior, error messages, timeline of when it started, and reproduction steps. Providing these upfront lets the investigation agent start with a complete symptom profile.

What happens when the debugger cannot find the root cause?▼

The agent returns an inconclusive report showing what was checked and eliminated. You can then continue investigating with a new agent, add more symptom context, or investigate manually.

Can I verify a fix before the debug session is closed?▼

Yes. When the agent reaches a human-verify checkpoint, it pauses and asks you to confirm the fix. If confirmed, the session is finalized and archived; if not, the agent returns to investigation.

Related Skills

Explore More Software Engineering Skills→
abrahamgreysonabrahamgreyson

gstack-openclaw-investigate

Trace code paths and validate hypotheses to identify root causes of software bugs.

Community
Advanced
PrestonigoPrestonigo

investigate

Guide structured debugging to identify root causes from symptoms, logs, and code paths.

Community
Advanced
JoaovsalesJoaovsales
1

debug

Guide bug investigation with reproduction, hypothesis ranking, and fix validation.

Community
Advanced
SethyPagnaSethyPagna

python-debugpy

Attach interactive debuggers to Python processes for breakpoint-driven tracing.

Community
Advanced