┌─────────────────────────────────────────────────────────────────────┐
│                    /consensus-review <scope>                        │
├─────────────────────────────────────────────────────────────────────┤
│  Phase 1: Parallel Initialization                                   │
│  ┌──────────────────────┐  ┌──────────────────────┐                │
│  │   Scope Detection    │  │   File Indexing      │                │
│  │   - Resolve files    │  │   - Index to contextd│                │
│  │   - Detect languages │  │   - Get token counts │                │
│  │   - Select agents    │  │   - Branch metadata  │                │
│  └──────────┬───────────┘  └──────────┬───────────┘                │
│             └──────────────┬──────────┘                            │
│                            ▼                                        │
│  Phase 2: Budget Calculation                                        │
│  ┌─────────────────────────────────────────────────────────────┐   │
│  │  total_tokens = sum(indexed_file_tokens)                    │   │
│  │  isolation_mode = total_tokens > 16K ? "branch" : "shared"  │   │
│  │  per_agent_budget = calculate_budget(total_tokens, agent)   │   │
│  └─────────────────────────────────────────────────────────────┘   │
│                            ▼                                        │
│  Phase 3: Agent Dispatch (parallel, with budgets)                  │
│  Phase 4: Synthesis (handles partial results)                      │
└─────────────────────────────────────────────────────────────────────┘

scale = min(4.0, 1.0 + total_tokens / 16384)
per_agent_budget = base_budget[agent] * scale

complexity_multiplier = {
  "SIMPLE": 0.75,    # Reduce budget for simple reviews
  "STANDARD": 1.0,   # Standard budget
  "COMPLEX": 1.5     # Increase budget for complex reviews
}

adjusted_budget = per_agent_budget * complexity_multiplier[tier]

For each agent:
  branch_create(
    description: "{agent} review of {scope}",
    budget: calculated_budget
  )
  → Agent executes in isolated context
  → branch_return(findings_json)

1. Determine complexity tier (from complexity-assessment or default STANDARD)
2. Select base agent set from triage matrix
3. Scan diff for override patterns
4. Merge override agents into base set (deduplicate)
5. Dispatch final agent set with calculated budgets

{
  "agent": "security-reviewer",
  "partial": true,
  "cutoff_reason": "budget",
  "files_reviewed": 8,
  "files_skipped": 4,
  "skipped_files": ["worker.go", "cache.go", ...],
  "findings": [...],
  "recommendation": "Re-run: /consensus-review worker.go cache.go"
}

total_tokens = sum(file.token_count for file in indexed_files)
scale = min(4.0, 1.0 + total_tokens / 16384)
complexity_multiplier = {"SIMPLE": 0.75, "STANDARD": 1.0, "COMPLEX": 1.5}
tier = complexity_assessment.tier or "STANDARD"  # Default if not available

for agent in selected_agents:
    agent.budget = base_budgets[agent] * scale * complexity_multiplier[tier]

if total_tokens > 16384:
    mode = "branch"  # Use contextd branch_create/branch_return
else:
    mode = "shared"  # Run in parent context

Task(
  subagent_type: "fs-dev:{agent}",
  prompt: |
    Review {scope}.
    Budget: {calculated_budget} tokens.
    See includes/consensus-review/progressive.md for degradation protocol.
  run_in_background: true
)

results = []
partial_agents = []

for agent in agents:
    output = TaskOutput(agent.task_id)
    results.append(output)

    if output.partial:
        partial_agents.append({
            "agent": output.agent,
            "files_reviewed": output.files_reviewed,
            "files_skipped": output.files_skipped
        })

# Calculate coverage per agent
# De-duplicate findings
# Check veto status
# Generate report

1. Dispatch all agents in parallel (no shared context)
2. Collect all agent outputs
3. Synthesis agent merges findings (no agent sees others' work)
4. Final report generated from merged findings

Round 1: All agents generate independent findings
Exchange: Findings (not rationale) shared across agents
Round 2: Agents update their findings based on others' discoveries
Synthesis: Final merge of updated findings

1. Collect all agent findings
2. Group findings by file + issue
3. Include in report only if >= 66% of reviewing agents flagged it
4. Exception: CRITICAL/HIGH security findings always included (veto applies)

/consensus-review <scope> --protocol aad    # All-Agents Drafting
/consensus-review <scope> --protocol ci     # Collective Improvement
/consensus-review <scope> --protocol vote   # Supermajority
/consensus-review <scope>                   # Default: Approval/Veto

| File         | Security | Vuln | Quality | Go  | Docs | Persona |
|--------------|----------|------|---------|-----|------|---------|
| handler.go   | Y        | Y    | Y       | Y   | -    | -       |
| worker.go    | SKIP     | Y    | Y       | Y   | -    | -       |
| README.md    | -        | -    | -       | -   | Y    | Y       |
| auth.go      | Y        | Y    | Y       | Y   | -    | Y       |

Coverage: 11/12 files fully covered
Gap: worker.go lacks security coverage (skipped due to budget)
→ Suggested: /consensus-review worker.go --agents security

repository_index(
  path: project_path,
  include_patterns: changed_files,
  branch: current_branch  # Auto-detected
)

branch_create(
  session_id: "review-{scope}-{timestamp}",
  description: "Security review of {scope}",
  budget: 16384
)

# ... agent executes ...

branch_return(
  message: "Security review complete: 2 CRITICAL, 1 HIGH"
)

memory_record(
  title: "Consensus Review: {scope}",
  content: "{summary}",
  tags: ["consensus-review", "{verdict}"]
)

═══ Consensus Review: src/auth/ ═══

Verdict: BLOCKED (2 vetoes)
⚠️  Partial results: security-reviewer hit budget limit (8/12 files)

┌────────────────┬─────────┬────┬────┬────┬────┬──────────┐
│ Agent          │ Verdict │ C  │ H  │ M  │ L  │ Coverage │
├────────────────┼─────────┼────┼────┼────┼────┼──────────┤
│ Security       │ VETO    │ 1  │ 0  │ 2  │ 0  │ 67% ⚠️   │
│ Vulnerability  │ OK      │ 0  │ 0  │ 1  │ 0  │ 100%     │
│ Code Quality   │ OK      │ 0  │ 1  │ 1  │ 0  │ 100%     │
└────────────────┴─────────┴────┴────┴────┴────┴──────────┘

Files not reviewed by security-reviewer: worker.go, cache.go
→ Consider: /consensus-review src/auth/worker.go src/auth/cache.go

═══ Consensus Review: src/api/ ═══

Verdict: APPROVED

┌────────────────┬─────────┬────┬────┬────┬────┬──────────┐
│ Agent          │ Verdict │ C  │ H  │ M  │ L  │ Coverage │
├────────────────┼─────────┼────┼────┼────┼────┼──────────┤
│ Security       │ OK      │ 0  │ 0  │ 1  │ 2  │ 100%     │
│ Vulnerability  │ OK      │ 0  │ 0  │ 0  │ 1  │ 100%     │
│ Code Quality   │ OK      │ 0  │ 0  │ 2  │ 3  │ 100%     │
└────────────────┴─────────┴────┴────┴────┴────┴──────────┘

Medium (3): ...
Low (6): ...