How do I install the S Debug skill?

Run npx skillmds add majiayu000/s-debug in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is S Debug safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with S Debug?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is S Debug free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published S Debug?

majiayu000 (@majiayu000) published this skill. Their other Agent Skills are listed on their SkillMD profile.

S Debug by majiayu000 — Agent Skill

Name: S Debug
Rating: 4.5 (4 reviews)
Author: majiayu000

S Debug

Diagnose and fix bugs using evidence-based investigation. Requires runtime evidence before fixes—never guess based on code alone. Covers hypothesis-driven debugging, instrumentation logging, Lua errors, taint issues, combat lockdown, and API failures. Triggers: error, bug, debug, fix, crash, taint, nil value, diagnose, hypothesis, runtime, evidence, instrumentation.

majiayu000 c5c7c12 2 files · 4.1 KB Updated Sep 9, 2026 567 repo stars

File contents

Debugging WoW Addons

Systematic debugging and error recovery for WoW addons.

Related Commands

c-debug - Reload loop workflow for finding and fixing issues
c-review - Full code review (includes debug step)

MCP Tools (Use These First)

MANDATORY: ALWAYS use MCP tools directly instead of the shell.

Task	MCP Tool
Get All Output	`addon.output(agent_mode=true)`
Lint Addon	`addon.lint(addon="MyAddon")`
Scan Deprecations	`addon.deprecations(addon="MyAddon")`
Queue Lua Eval	`lua.queue(code=["GetMoney()"])`
Get Eval Results	`lua.results()`

Capabilities

Evidence-Based Debugging — Hypothesis-driven investigation with runtime instrumentation and log proof
Error Analysis — Parse Lua errors, identify root cause in stack traces
Taint Investigation — Track secure/insecure code interaction and "Action blocked" issues
Combat Issues — Debug lockdown-related failures and protected frame issues
API Failures — Handle deprecated or changed APIs (Midnight 12.0 prep)

Routing Logic

Request type	Load reference
Evidence-based debugging, hypothesis-driven	references/evidence-based-debugging.md
Lua errors, nil values	references/error-patterns.md
Debug workflow, isolation	references/debugging-strategies.md
Error tracking (BugGrabber)	../../docs/integration/errors.md
Troubleshooting guide	../../docs/integration/troubleshooting.md
Structured logging	../../docs/integration/console.md
Frame inspection	../../docs/integration/inspect.md

Debug Output Best Practice

CRITICAL: Use MechanicLib:Log() instead of print() for all debug output.

Feature	`print()`	`MechanicLib:Log()`
Agent access	❌ Requires screenshot	✅ `addon.output` retrieves directly
Filtering	❌ None	✅ Source + category filters
Copyable	❌ No	✅ Yes, via Console export

-- ✅ Correct: Agent can see this in addon.output
local MechanicLib = LibStub("MechanicLib-1.0", true)
if MechanicLib then
    MechanicLib:Log("MyAddon", "Debug: value=" .. tostring(val), MechanicLib.Categories.CORE)
end

-- ❌ Avoid: Spams chat, requires screenshot
print("[MyAddon] Debug: value=" .. tostring(val))

Quick Reference

Get Addon Data (Compressed for AI)

Ask user to /reload and confirm, then:

addon.output(agent_mode=true)

Common Error Patterns

attempt to index nil value: API returned nil, check if unit exists or data is loaded.
Action blocked by Blizzard: You tried to call a protected function in combat.
Interface action failed: Taint has spread to a secure UI component.

Systematic Workflow

Gather Evidence: Ask user to /reload, wait for confirmation, then addon.output(agent_mode=true)
Isolate: Can you reproduce with minimal code?
Hypothesis: "If X then Y because Z"
Fix & Validate: Apply minimal fix, ask user to /reload and confirm, then verify with addon.output.

---name: s-debugdescription: Diagnose and fix bugs using evidence-based investigation. Requires runtime evidence before fixes—never guess based on code alone. Covers hypothesis-driven debugging, instrumentation logging, Lua errors, taint issues, combat lockdown, and API failures. Triggers: error, bug, debug, fix, crash, taint, nil value, diagnose, hypothesis, runtime, evidence, instrumentation.--- # Debugging WoW Addons Systematic debugging and error recovery for WoW addons. ## Related Commands - [c-debug](../../commands/c-debug.md) - Reload loop workflow for finding and fixing issues - [c-review](../../commands/c-review.md) - Full code review (includes debug step) ## MCP Tools (Use These First) > **MANDATORY**: ALWAYS use MCP tools directly instead of the shell. | Task | MCP Tool | |------|----------| | Get All Output | `addon.output(agent_mode=true)` | | Lint Addon | `addon.lint(addon="MyAddon")` | | Scan Deprecations | `addon.deprecations(addon="MyAddon")` | | Queue Lua Eval | `lua.queue(code=["GetMoney()"])` | | Get Eval Results | `lua.results()` | ## Capabilities 1. **Evidence-Based Debugging** — Hypothesis-driven investigation with runtime instrumentation and log proof 2. **Error Analysis** — Parse Lua errors, identify root cause in stack traces 3. **Taint Investigation** — Track secure/insecure code interaction and "Action blocked" issues 4. **Combat Issues** — Debug lockdown-related failures and protected frame issues 5. **API Failures** — Handle deprecated or changed APIs (Midnight 12.0 prep) ## Routing Logic | Request type | Load reference | |--------------|----------------| | Evidence-based debugging, hypothesis-driven | [references/evidence-based-debugging.md](references/evidence-based-debugging.md) | | Lua errors, nil values | [references/error-patterns.md](references/error-patterns.md) | | Debug workflow, isolation | [references/debugging-strategies.md](references/debugging-strategies.md) | | Error tracking (BugGrabber) | [../../docs/integration/errors.md](../../docs/integration/errors.md) | | Troubleshooting guide | [../../docs/integration/troubleshooting.md](../../docs/integration/troubleshooting.md) | | Structured logging | [../../docs/integration/console.md](../../docs/integration/console.md) | | Frame inspection | [../../docs/integration/inspect.md](../../docs/integration/inspect.md) | ## Debug Output Best Practice > **CRITICAL**: Use `MechanicLib:Log()` instead of `print()` for all debug output. | Feature | `print()` | `MechanicLib:Log()` | |---------|-----------|---------------------| | Agent access | ❌ Requires screenshot | ✅ `addon.output` retrieves directly | | Filtering | ❌ None | ✅ Source + category filters | | Copyable | ❌ No | ✅ Yes, via Console export | ```lua -- ✅ Correct: Agent can see this in addon.output local MechanicLib = LibStub("MechanicLib-1.0", true) if MechanicLib then MechanicLib:Log("MyAddon", "Debug: value=" .. tostring(val), MechanicLib.Categories.CORE) end -- ❌ Avoid: Spams chat, requires screenshot print("[MyAddon] Debug: value=" .. tostring(val)) ``` ## Quick Reference ### Get Addon Data (Compressed for AI) **Ask** user to `/reload` and confirm, then: ```bash addon.output(agent_mode=true) ``` ### Common Error Patterns - `attempt to index nil value`: API returned nil, check if unit exists or data is loaded. - `Action blocked by Blizzard`: You tried to call a protected function in combat. - `Interface action failed`: Taint has spread to a secure UI component. ### Systematic Workflow 1. **Gather Evidence**: **Ask** user to `/reload`, wait for confirmation, then `addon.output(agent_mode=true)` 2. **Isolate**: Can you reproduce with minimal code? 3. **Hypothesis**: "If X then Y because Z" 4. **Fix & Validate**: Apply minimal fix, **ask** user to `/reload` and confirm, then verify with `addon.output`.

Frequently asked questions