Stats
Actions
Tags
From augments
Use when you have two or more genuinely independent pieces of work — separate failing tests, unrelated bugs, parallel research threads — that share no files, no state, and no ordering. Fan them out to concurrent agents, each with its own scope and deliverable, then reconcile. Skip when tasks depend on each other or share a file (sequence them with executing-plans), or when the work is quick enough inline.
How this skill is triggered — by the user, by Claude, or both
Slash command
/augments:dispatching-parallel-agentsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Run independent work concurrently instead of in series. The win is wall-clock; the risk is collision — so this applies only when the pieces genuinely don't touch each other.
Run independent work concurrently instead of in series. The win is wall-clock; the risk is collision — so this applies only when the pieces genuinely don't touch each other.
executing-plans), when one needs another's output, or when the whole thing is quick inline.Check every pair; if any fails, group them into one agent or sequence them instead:
using-git-worktrees).Each agent starts cold — hand it everything, never your session history:
Pass bulky context — a diff, a spec, a log — as a file path the agent reads, not pasted text: a paste sits in the most expensive context for the whole run, a path costs nothing until opened.
When agents return: read each deliverable, integrate them yourself, then run the combined check — the tests across all the changes together. N green agents do not prove the whole is green. Surface any agent that reported blocked or out-of-scope rather than papering over it.
using-git-worktrees).npx claudepluginhub njoyimpeguy/augments --plugin augmentsGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.