Parallel Agents

When you need multiple specialized perspectives on a codebase, this orchestrates Claude's 17 domain-specific agents (security auditors, frontend specialists, test engineers, etc.) to work in parallel or sequence. You invoke them conversationally, like "use the security-auditor agent to review authentication, then have the test-engineer check coverage." The real win is context passing between agents and getting a single synthesized report instead of scattered outputs. Best for comprehensive analysis where you need security, performance, and architecture viewpoints on the same code. Overkill for simple fixes, but solid for feature reviews or full audits where one agent's perspective isn't enough.