Overview

Dispatching Parallel Agents assigns one agent per independent problem domain and lets them work concurrently. The core principle is to avoid shared state and sequential dependencies by keeping problems separate. This approach speeds up triage when multiple subsystems fail independently.

How This Skill Works

Identify independent domains and dispatch agents in parallel using ctx.parallel.all. Each agent operates on its own domain with no shared state. After solutions are produced, check for conflicts and run the full test suite to verify integration.

When to Use It

• 3+ test files failing with different root causes
• Multiple subsystems broken independently
• Each problem understood without context from others
• No shared state between investigations

Quick Start

1. Step 1: Identify independent domains
2. Step 2: Dispatch agents in parallel (ctx.parallel.all)
3. Step 3: Check for conflicts and run the full test suite to verify integration

Best Practices

• Map problems to independent domains before dispatch
• Ensure no shared state exists between domains
• Dispatch agents in parallel using ctx.parallel.all
• Check for conflicts between solutions after parallel runs
• Run the full test suite to verify integration

Example Use Cases

• 3+ test files failing with different root causes
• Multiple subsystems broken independently
• Each problem understood without context from others
• No shared state between investigations
• After parallel runs, conflicts are identified and validated with integration tests

Frequently Asked Questions