dispatching-parallel-agents

# Dispatching Parallel Agents ## Overview When you have multiple unrelated failures (different test files, different subsystems, different bugs), investigating them sequentially wastes time. Each investigation is independent and can happen in parallel. **Core principle:** Dispatch one agent per independent problem domain. ## When to Use Use when: 3+ test files failing with different root causes, multiple subsystems broken independently, each problem can be understood without context from others. Don't use when: Failures are related, need to understand full system state, agents would interfere. ## The Pattern 1. Identify Independent Domains - Group failures by what's broken 2. Create Focused Agent Tasks - Each agent gets specific scope, clear goal, constraints 3. Dispatch in Parallel 4. Review and Integrate ## Agent Prompt Structure Good prompts are: Focused (one problem), Self-contained (all context), Specific about output. ## Common Mistakes ❌ Too broad - "Fix all the tests" ❌ No context - agent doesn't know where ❌ No constraints - agent might refactor everything ❌ Vague output - you don't know what changed ## Verification After agents return: Review each summary, Check for conflicts, Run full suite, Spot check.