Parallel Debugging

# Hypothesis Testing Reference

Task templates, evidence formats, and arbitration decision trees for parallel debugging.

## Hypothesis Task Template

```markdown
## Hypothesis Investigation: {Hypothesis Title}

### Hypothesis Statement

{Clear, falsifiable statement about the root cause}

### Failure Mode Category

{Logic Error | Data Issue | State Problem | Integration Failure | Resource Issue | Environment}

### Investigation Scope

- Files to examine: {file list or directory}
- Related tests: {test files}
- Git history: {relevant date range or commits}

### Evidence Criteria

**Confirming evidence** (if I find these, hypothesis is supported):

1. {Observable condition 1}
2. {Observable condition 2}

**Falsifying evidence** (if I find these, hypothesis is wrong):

1. {Observable condition 1}
2. {Observable condition 2}

### Report Format

- Confidence: High/Medium/Low
- Evidence: list with file:line citations
- Causal chain: step-by-step from cause to symptom
- Recommended fix: if confirmed
```

## Evidence Report Template

```markdown
## Investigation Report: {Hypothesis Title}

### Verdict: {Confirmed | Falsified | Inconclusive}

### Confidence: {High (>80%) | Medium (50-80%) | Low (<50%)}

### Confirming Evidence

1. `src/api/users.ts:47` — {description of what was found}
2. `src/middleware/auth.ts:23` — {description}

### Contradicting Evidence

1. `tests/api/users.test.ts:112` — {description of what contradicts}

### Causal Chain (if confirmed)

1. {First cause} →
2. {Intermediate effect} →
3. {Observable symptom}

### Recommended Fix

{Specific code change with location}

### Additional Notes

{Anything discovered that may be relevant to other hypotheses}
```

## Arbitration Decision Tree

```
All investigators reported?
├── NO → Wait for remaining reports
└── YES → Count confirmed hypotheses
          ├── 0 confirmed
          │   ├── Any medium confidence? → Investigate further
          │   └── All low/falsified? → Generate new hypotheses
          ├── 1 confirmed
          │   └── High confidence?
          │       ├── YES → Declare root cause, propose fix
          │       └── NO → Flag as likely cause, recommend verification
          └── 2+ confirmed
              └── Are they related?
                  ├── YES → Compound issue (multiple contributing causes)
                  └── NO → Rank by confidence, declare highest as primary
```

## Common Hypothesis Patterns by Error Type

### "500 Internal Server Error"

1. Unhandled exception in request handler (Logic Error)
2. Database connection failure (Resource Issue)
3. Missing environment variable (Environment)

### "Race condition / intermittent failure"

1. Shared state mutation without locking (State Problem)
2. Async operation ordering assumption (Logic Error)
3. Cache staleness window (State Problem)

### "Works locally, fails in production"

1. Environment variable mismatch (Environment)
2. Different dependency version (Environment)
3. Resource limits (memory, connections) (Resource Issue)

### "Regression after deploy"

1. New code introduced bug (Logic Error)
2. Configuration change (Integration Failure)
3. Database migration issue (Data Issue)


---
name: parallel-debugging
description: Debug complex issues using competing hypotheses with parallel investigation, evidence collection, and root cause arbitration. Use this skill when debugging bugs with multiple potential causes, performing root cause analysis, or organizing parallel investigation workflows.
version: 1.0.2
---

# Parallel Debugging

Framework for debugging complex issues using the Analysis of Competing Hypotheses (ACH) methodology with parallel agent investigation.

## When to Use This Skill

- Bug has multiple plausible root causes
- Initial debugging attempts haven't identified the issue
- Issue spans multiple modules or components
- Need systematic root cause analysis with evidence
- Want to avoid confirmation bias in debugging

## Hypothesis Generation Framework

Generate hypotheses acro