dispatching-parallel-agentslisted

<!-- Adapted from superpowers (https://github.com/obra/superpowers), MIT (c) Jesse Vincent. --> # Dispatching Parallel Agents ## Overview You delegate to specialized agents with isolated context, crafting exactly the instructions and context each needs (never your session history). This keeps them focused and preserves your context for coordination. When you have multiple unrelated problems (different test files, subsystems, bugs), investigating sequentially wastes time. Each is independent and can happen in parallel. **Core principle:** Dispatch one agent per independent problem domain. Let them work concurrently. ## Fits in the pipeline Powers the **coding swarm in Stage 6 (Build)** — FE/BE/data/infra agents working independent modules at once — and parallel defect triage feeding the **Stage 7 per-feature QA gate**. Pair with `using-git-worktrees` so parallel agents don't collide on the same files, and `subagent-driven-development` for per-task review within a domain. Priority: **user > skills > default**; `_shared/vibegod-principles.md` apply. ## When to Use ```dot digraph when_to_use { "Multiple failures/tasks?" [shape=diamond]; "Are they independent?" [shape=diamond]; "Single agent investigates all" [shape=box]; "Can they work in parallel?" [shape=diamond]; "Sequential agents" [shape=box]; "Parallel dispatch" [shape=box]; "Multiple failures/tasks?" -> "Are they independent?" [label="yes"]; "Are they independent?" -> "Single agent