---
name: 'step-05-consolidate-findings'
description: 'Merge and deduplicate findings from both review phases'
---

# Step 5: Consolidate Findings

**Goal:** Merge findings from adversarial review (Phase 1) and context-aware review (Phase 2), deduplicate, and present unified findings table.

---

## AVAILABLE STATE

From previous steps:

- `{story_path}`, `{story_key}`
- `{asymmetric_findings}` - Findings from Phase 1 (step 2 - adversarial review)
- `{context_aware_findings}` - Findings from Phase 2 (step 4 - context-aware review)

---

## STATE VARIABLE (capture now)

- `{consolidated_findings}` - Merged, deduplicated findings

---

## EXECUTION SEQUENCE

### 1. Merge All Findings

Combine both finding lists:

```
all_findings = {context_aware_findings} + {asymmetric_findings}
```

### 2. Deduplicate Findings

Identify duplicates (same underlying issue found by both phases):

**Duplicate Detection Criteria:**

- Same file + same line range
- Same issue type (e.g., both about error handling in same function)
- Overlapping descriptions

**Resolution Rule:**

Keep the MORE DETAILED version:

- If context-aware finding has AC reference → keep that
- If adversarial finding has better technical detail → keep that
- When in doubt, keep context-aware (has more context)

Note which findings were merged (for transparency in the summary).

### 3. Normalize Severity

Apply consistent severity scale (Critical, High, Medium, Low).

### 4. Filter Noise

Review adversarial findings marked as Noise:

- If clearly false positive (e.g., style preference, not actual issue) → exclude
- If questionable → keep with Undecided validity
- If context reveals it's actually valid → upgrade to Real

**Do NOT filter:**

- Any Critical or High severity
- Any context-aware findings (they have story context)

### 5. Sort and Number Findings

Sort by severity (Critical → High → Medium → Low), then assign IDs: F1, F2, F3, etc.

Build `{consolidated_findings}`:

```markdown
| ID | Severity | Source | Description | Location |
|----|----------|--------|-------------|----------|
| F1 | Critical | task-audit | Task 3 marked [x] but not implemented | src/auth.ts |
| F2 | High | ac-validation | AC2 partially implemented | src/api/*.ts |
| F3 | High | adversarial | Missing error handling in API calls | src/api/client.ts:45 |
| F4 | Medium | git-discrepancy | File changed but not in story | src/utils.ts |
| F5 | Low | adversarial | Magic number should be constant | src/config.ts:12 |
```

### 6. Present Consolidated Findings

```markdown
**Consolidated Code Review Findings**

**Story:** {story_key}

**Summary:**
- Total findings: {count}
- Critical: {count}
- High: {count}
- Medium: {count}
- Low: {count}

**Deduplication:** {merged_count} duplicate findings merged

---

## Findings by Severity

### Critical (Must Fix)
{list critical findings with full details}

### High (Should Fix)
{list high findings with full details}

### Medium (Consider Fixing)
{list medium findings}

### Low (Nice to Fix)
{list low findings}

---

**Phase Sources:**
- Adversarial (Phase 1): {count} findings
- Context-Aware (Phase 2): {count} findings
```

---

## NEXT STEP DIRECTIVE

**CRITICAL:** When this step completes, explicitly state:

"**NEXT:** Loading `step-06-resolve-and-update.md`"

---

## SUCCESS METRICS

- All findings merged from both phases
- Duplicates identified and resolved (kept more detailed)
- Severity normalized consistently
- Noise filtered appropriately (but not excessively)
- Consolidated table created
- `{consolidated_findings}` populated
- Summary presented to user
- Explicit NEXT directive provided

## FAILURE MODES

- Missing findings from either phase
- Not detecting duplicates (double-counting issues)
- Inconsistent severity assignment
- Filtering real issues as noise
- Not storing consolidated findings
- No explicit NEXT directive at step completion