Dispatching Parallel Agents

When you've got multiple independent failures after a refactor, this is about spinning up separate agents to tackle each problem domain in parallel instead of debugging sequentially. Each agent gets a tightly scoped prompt with specific context, clear constraints, and an instruction to report back what they found. The key insight is recognizing when problems are truly independent (abort tests, batch completion, race conditions) versus related. In the documented example, three agents fixed six test failures concurrently across different files with zero conflicts. The pattern works when there's no shared state and you can clearly partition the work, but it requires discipline in crafting focused prompts and verifying the results don't step on each other.

GitHub stars: 492
last updated: Jun 11, 2026

View on GitHub

Install

$ npx -y skills add ganyuanran/aegis --skill dispatching-parallel-agents --agent claude-code

Installs into .claude/skills of the current project.

Files

Dispatching Parallel Agents

Purpose

Use parallel children for two or more ad-hoc, bounded tasks only when their independence is already credible and concurrency is worth the coordination. Parallelism changes elapsed time; it does not relax evidence, ownership, or completion requirements.

Dispatch Gate

Dispatch only when all are true:

there are 2+ distinct tasks or evidence questions;
one task's result is not needed to define or start another;
the tasks do not require overlapping writes or the same mutable resource;
a shared root cause is unlikely based on current evidence;
each result can be returned with bounded evidence and unresolved unknowns;
available concurrency and expected work justify coordination cost.

Keep work inline when decomposition is still exploratory, failures may share a root cause, full-system context is required, or tasks are too small to repay the handoff. A written implementation plan routes to aegis:subagent-driven-development, not this ad-hoc workflow.