MEGA SWARM — Parallel Codebase Audit

Maximum coverage. Every angle. One report.

Scope: $1 (default: full | full|src|tests|config|security) Focus: $2 (optional focus area) Mode: $3 (default: full | full|quick|focused) Quick: $4 (default: false — same as mode=quick)

TEAM ARCHITECTURE

SWARM LEAD (You — Orchestrator)
│
├─ Full mode (12 agents) ─────────────────────
│  ├── arch-auditor        ├── api-auditor
│  ├── security-auditor    ├── dependency-auditor
│  ├── perf-auditor        ├── config-auditor
│  ├── test-auditor        ├── docs-auditor
│  ├── quality-auditor     ├── consistency-auditor
│  ├── error-auditor       └── bug-hunter
│
├─ Quick mode (6 agents) ─────────────────────
│  ├── arch-auditor    ├── security-auditor
│  ├── perf-auditor    ├── test-auditor
│  ├── quality-auditor └── bug-hunter
│
├─ Focused mode (8 agents) ───────────────────
│  └── $2-related auditors + adjacent concerns
│
└── GATE → >=80% complete → SYNTHESIZE

<CRITICAL_EXECUTION_REQUIREMENT>

Launch ALL agents for selected mode in ONE message. NOTHING ELSE.

• •Full ($3=full or unspecified): 12 agents
• •Quick ($3=quick or $4=true): 6 agents
• •Focused ($3=focused): 8 agents relevant to $2

Wait for completion. Synthesize. Report.

YOUR NEXT MESSAGE: Task tool calls for selected mode. NOTHING ELSE.

</CRITICAL_EXECUTION_REQUIREMENT>

AUDITORS

Each scans SCOPE=$1 through its lens. Output: issues with severity P0-P3.

arch-auditor

subagent: metacognitive-guard:arch-reviewer | model: opus

AUDIT Architecture. Scope: $1 | Focus: $2 SOLID violations? Coupling? Scalability? Design issues? Output: Issues with P0-P3 severity

security-auditor

subagent: feature-dev:code-reviewer

AUDIT Security. Scope: $1 | Focus: $2 Injection? Auth issues? Secrets exposed? Input validation? OWASP Top 10? Output: Issues with severity

perf-auditor

subagent: feature-dev:code-explorer

AUDIT Performance. Scope: $1 | Focus: $2 N+1 queries? Memory leaks? Allocations? Blocking calls? Cache misses? Output: Issues with severity

test-auditor

subagent: feature-dev:code-reviewer

AUDIT Test Quality. Scope: $1 | Focus: $2 Coverage gaps? Flaky tests? Missing edge cases? Integration gaps? Output: Issues with severity

quality-auditor

subagent: feature-dev:code-reviewer

AUDIT Code Quality. Scope: $1 | Focus: $2 Dead code? Duplication? Cyclomatic complexity? Magic numbers? Naming? Output: Issues with severity

bug-hunter

subagent: deep-debugger | model: opus

HUNT Active Bugs. Scope: $1 | Focus: $2 Null refs? Race conditions? Off-by-one? Resource leaks? Logic errors? Output: Potential bugs with severity

error-auditor ← full/focused only

subagent: deep-debugger

AUDIT Error Handling. Scope: $1 | Focus: $2 Swallowed exceptions? Missing handlers? Poor messages? Recovery gaps? Output: Issues with severity

api-auditor ← full/focused only

subagent: feature-dev:code-explorer

AUDIT API Contracts. Scope: $1 | Focus: $2 Breaking changes? Version compat? Doc accuracy? Response consistency? Output: Issues with severity

dependency-auditor ← full/focused only

subagent: Explore

AUDIT Dependencies. Scope: $1 | Focus: $2 Outdated? Vulnerabilities? License issues? Unnecessary? Conflicts? Output: Issues with severity

config-auditor ← full/focused only

subagent: Explore

AUDIT Configuration. Scope: $1 | Focus: $2 Hardcoded values? Missing env vars? Validation? Secrets management? Output: Issues with severity

docs-auditor ← full/focused only

subagent: Explore

AUDIT Documentation. Scope: $1 | Focus: $2 Outdated? Missing? Code comments? README accuracy? Output: Issues with severity

consistency-auditor ← full/focused only

subagent: feature-dev:code-reviewer

AUDIT Consistency. Scope: $1 | Focus: $2 Naming conventions? Code style? Pattern inconsistencies? File organization? Output: Issues with severity

GATE: Audit Complete

AUDIT GATE:
+------------------------------------------------------------+
| Mode: $3 | Agents: [X/Y] completed                        |
+------------------------------------------------------------+
| >=80% → SYNTHESIZE                                          |
| <80% → REPORT PARTIAL + offer retry                         |
+------------------------------------------------------------+

FINAL REPORT

+====================================================================+
|                    MEGA SWARM REPORT                                |
+====================================================================+
| Mode: $3 | Agents: [count] | Scope: $1                             |
+--------------------------------------------------------------------+
| P0 Critical: [n] | P1 High:   [n]                                  |
| P2 Medium:   [n] | P3 Low:    [n]                                  |
+--------------------------------------------------------------------+
| Security:    [n] | Performance: [n] | Architecture: [n]             |
| Tests:       [n] | Quality:     [n] | Bugs:         [n]             |
+====================================================================+

P0 Issues (Fix Immediately)

Recommended Fix Order

1. •[Most critical]
2. •[Second most critical]

Next: /fix-pipeline "[P0 issue]"