Skip to content
All Skills

Troubleshoot

Systematic debugging with 5-step loop for issues in spaces/[project]/. Use when encountering bugs, test failures, or unexpected behavior.

Software Engineering|v1|Updated 7/14/2026|GitHub source
MCP get_skill({ skillId: "troubleshoot-6e26ab7e" })

Use this skill with your agent

Create a free account and connect via MCP

Get Started Free
# /troubleshoot

Systematic debugging using a 5-step loop: Research → Hypothesize → Implement → Test → Document.

## Usage

```bash
/troubleshoot yourbench "tests failing after auth changes"
/troubleshoot yourbench 001       # Debug in context of issue
/troubleshoot coordinatr          # General debugging session
```

## The 5-Step Loop

```
┌─────────────────────────────────────────────┐
│                                             │
│  1. Research → 2. Hypothesize → 3. Implement│
│       ↑                              ↓      │
│       │                              │      │
│       └──────── 5. Document ← 4. Test ──────│
│                     │                       │
│                     ↓                       │
│              (if not fixed, repeat)         │
│                                             │
└─────────────────────────────────────────────┘
```

## Step Details

### Step 1: Research

**Before guessing, understand:**

```bash
# Check existing research
Glob: resources/research/*.md
Glob: ideas/[project]/notes/research/*.md

# Search for similar issues in codebase
Grep: spaces/[project]/ for error message

# Check library docs via Context7
# WebSearch for error messages, known issues
```

Spawn research-specialist agent for:
- Unfamiliar error messages
- Library/framework behavior
- Best practices for the pattern

### Step 2: Hypothesize

**Form ONE hypothesis:**

```markdown
Hypothesis: Query fires before auth state is set

Evidence:
- isLoading is true when query executes
- Error occurs only on initial load
- Works after manual refresh

Debug Plan:
1. Add console.log before query
2. Check auth state timing
3. Verify order of operations
```

**Only one hypothesis at a time.** Multiple theories = confusion.

### Step 3: Implement

**Apply fix + add debugging:**

```typescript
// Add liberal debug logging
console.log('[Auth] State before query:', authState);
console.log('[Query] Executing with user:', user?.id);

// Implement the fix
if (!authState.isReady) {
  console.log('[Query] Waiting for auth...');
  return;
}
```

### Step 4: Test

**Validate the fix:**

```bash
cd spaces/[project]

# Run specific tests
npm test -- --grep "auth"

# Run full suite
npm test

# Manual verification if needed
npm run dev
```

**NEVER claim "fixed" without tests passing.**

### Step 5: Document

**Update WORKLOG with findings:**

```markdown
## YYYY-MM-DD - Troubleshooting Loop 1

**Hypothesis**: Query fires before auth state is set

**Debug findings**:
- isLoading was true when query executed
- Auth state not propagating to component
- Race condition between auth init and query

**Implementation**: Added auth state check before query

**Result**: ✓ Fixed - 47/47 tests passing

**Gotcha**: Auth state needs explicit ready check, not just truthy
```

### If Not Fixed

1. **Rollback changes**: `git checkout -- .`
2. **Document what you learned**
3. **Return to Step 1** with new information

## Execution Modes

### Mode 1: During TASK Implementation

```bash
/troubleshoot yourbench 001 "tests failing"
```

1. Read: TASK.md, PLAN.md, WORKLOG.md for context
2. Execute 5-step loop
3. Document in issue's WORKLOG.md
4. Continue with `/implement` when fixed

### Mode 2: Standalone Debugging

```bash
/troubleshoot yourbench "login broken on Safari"
```

1. Investigate without issue context
2. Document findings
3. Suggest creating BUG issue if significant

## Key Rules

1. **Research BEFORE guessing**
2. **ONE hypothesis at a time**
3. **Liberal debug logging** - `console.log('[Component] State:', data)`
4. **NEVER claim "fixed"** without tests passing
5. **Rollback on failure** before next attempt
6. **Document everything** for future reference

## Workflow

```
/implement → (issue occurs) → /troubleshoot → /worklog → /implement
```

**After fixing:**
- Add WORKLOG entry with findings
- Continue implementation
- Or create BUG issue if recurring problem
#personal-productivity#daily-routine#weekly-review#life-planning#personal-knowledge#debugginggitnodejs

Related Skills

More skills in Software Engineering

Accessibility Standards

Comprehensive web accessibility standards based on WCAG 2.2 AA, with 38+ anti-patterns, legal enforcement context (EAA, ADA Title II), WAI-ARIA patterns, and framework-specific fixes for modern web frameworks and libraries.

#github-copilot#accessibilityMIT

Accord

Authoring unified specification packages across Business/Development/Design teams via staged elaboration (L0 Vision → L1 Requirements → L2 Team Detail → L3 Acceptance Criteria). No code. Use when authoring cross-team specs, building L0-L3 packages, or aligning Biz/Dev/Design on a single source of truth.

#broad-capability#developmentMIT

Acquire Codebase Knowledge

Use this skill when the user explicitly asks to map, document, or onboard into an existing codebase. Trigger for prompts like "map this codebase", "document this architecture", "onboard me to this repo", or "create codebase docs". Do not trigger for routine feature implementation, bug fixes, or narrow code edits unless the user asks for repository-level discovery.

#github-copilot#documentationMIT

Acreadiness Assess

Run the AgentRC readiness assessment on the current repository and produce a static HTML dashboard at reports/index.html. Wraps `npx github:microsoft/agentrc readiness` and hands off rendering to the @ai-readiness-reporter custom agent. Supports policies (--policy) for org-specific scoring. Use when asked to assess, audit, or score the AI readiness of a repo.

#github-copilot#planningMIT

Acreadiness Generate Instructions

Generate tailored AI agent instruction files via AgentRC instructions command. Produces .github/copilot-instructions.md (default, recommended for Copilot in VS Code) plus optional per-area .instructions.md files with applyTo globs for monorepos. Use after running /acreadiness-assess to close gaps in the AI Tooling pillar.

#github-copilot#skillMIT

Acreadiness Policy

Help the user pick, write, or apply an AgentRC policy. Policies customise readiness scoring by disabling irrelevant checks, overriding impact/level, setting pass-rate thresholds, or chaining org baselines with team overrides. Use when the user asks about strict mode, AI-only scoring, custom weights, CI gating, or wants org-wide standardisation.

#github-copilot#planningMIT